linux/fs/nilfs2/mdt.c

650 lines
16 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0+
/*
* Meta data file for NILFS
*
* Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation.
*
* Written by Ryusuke Konishi.
*/
#include <linux/buffer_head.h>
#include <linux/mpage.h>
#include <linux/mm.h>
#include <linux/writeback.h>
#include <linux/backing-dev.h>
#include <linux/swap.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 "nilfs.h"
#include "btnode.h"
#include "segment.h"
#include "page.h"
#include "mdt.h"
#include "alloc.h" /* nilfs_palloc_destroy_cache() */
#include <trace/events/nilfs2.h>
#define NILFS_MDT_MAX_RA_BLOCKS (16 - 1)
static int
nilfs_mdt_insert_new_block(struct inode *inode, unsigned long block,
struct buffer_head *bh,
void (*init_block)(struct inode *,
struct buffer_head *, void *))
{
struct nilfs_inode_info *ii = NILFS_I(inode);
void *kaddr;
int ret;
/* Caller exclude read accesses using page lock */
/* set_buffer_new(bh); */
bh->b_blocknr = 0;
ret = nilfs_bmap_insert(ii->i_bmap, block, (unsigned long)bh);
if (unlikely(ret))
return ret;
set_buffer_mapped(bh);
kaddr = kmap_atomic(bh->b_page);
memset(kaddr + bh_offset(bh), 0, i_blocksize(inode));
if (init_block)
init_block(inode, bh, kaddr);
flush_dcache_page(bh->b_page);
kunmap_atomic(kaddr);
set_buffer_uptodate(bh);
mark_buffer_dirty(bh);
nilfs_mdt_mark_dirty(inode);
trace_nilfs2_mdt_insert_new_block(inode, inode->i_ino, block);
return 0;
}
static int nilfs_mdt_create_block(struct inode *inode, unsigned long block,
struct buffer_head **out_bh,
void (*init_block)(struct inode *,
struct buffer_head *,
void *))
{
struct super_block *sb = inode->i_sb;
struct nilfs_transaction_info ti;
struct buffer_head *bh;
int err;
nilfs_transaction_begin(sb, &ti, 0);
err = -ENOMEM;
bh = nilfs_grab_buffer(inode, inode->i_mapping, block, 0);
if (unlikely(!bh))
goto failed_unlock;
err = -EEXIST;
if (buffer_uptodate(bh))
goto failed_bh;
wait_on_buffer(bh);
if (buffer_uptodate(bh))
goto failed_bh;
bh->b_bdev = sb->s_bdev;
err = nilfs_mdt_insert_new_block(inode, block, bh, init_block);
if (likely(!err)) {
get_bh(bh);
*out_bh = bh;
}
failed_bh:
unlock_page(bh->b_page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(bh->b_page);
brelse(bh);
failed_unlock:
if (likely(!err))
err = nilfs_transaction_commit(sb);
else
nilfs_transaction_abort(sb);
return err;
}
static int
nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff,
int mode, int mode_flags, struct buffer_head **out_bh)
{
struct buffer_head *bh;
__u64 blknum = 0;
int ret = -ENOMEM;
bh = nilfs_grab_buffer(inode, inode->i_mapping, blkoff, 0);
if (unlikely(!bh))
goto failed;
ret = -EEXIST; /* internal code */
if (buffer_uptodate(bh))
goto out;
if (mode_flags & REQ_RAHEAD) {
if (!trylock_buffer(bh)) {
ret = -EBUSY;
goto failed_bh;
}
} else /* mode == READ */
lock_buffer(bh);
if (buffer_uptodate(bh)) {
unlock_buffer(bh);
goto out;
}
ret = nilfs_bmap_lookup(NILFS_I(inode)->i_bmap, blkoff, &blknum);
if (unlikely(ret)) {
unlock_buffer(bh);
goto failed_bh;
}
map_bh(bh, inode->i_sb, (sector_t)blknum);
bh->b_end_io = end_buffer_read_sync;
get_bh(bh);
submit_bh(mode, mode_flags, bh);
ret = 0;
trace_nilfs2_mdt_submit_block(inode, inode->i_ino, blkoff, mode);
out:
get_bh(bh);
*out_bh = bh;
failed_bh:
unlock_page(bh->b_page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(bh->b_page);
brelse(bh);
failed:
return ret;
}
static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
int readahead, struct buffer_head **out_bh)
{
struct buffer_head *first_bh, *bh;
unsigned long blkoff;
int i, nr_ra_blocks = NILFS_MDT_MAX_RA_BLOCKS;
int err;
err = nilfs_mdt_submit_block(inode, block, REQ_OP_READ, 0, &first_bh);
if (err == -EEXIST) /* internal code */
goto out;
if (unlikely(err))
goto failed;
if (readahead) {
blkoff = block + 1;
for (i = 0; i < nr_ra_blocks; i++, blkoff++) {
err = nilfs_mdt_submit_block(inode, blkoff, REQ_OP_READ,
REQ_RAHEAD, &bh);
if (likely(!err || err == -EEXIST))
brelse(bh);
else if (err != -EBUSY)
break;
/* abort readahead if bmap lookup failed */
if (!buffer_locked(first_bh))
goto out_no_wait;
}
}
wait_on_buffer(first_bh);
out_no_wait:
err = -EIO;
if (!buffer_uptodate(first_bh)) {
nilfs_err(inode->i_sb,
"I/O error reading meta-data file (ino=%lu, block-offset=%lu)",
inode->i_ino, block);
goto failed_bh;
}
out:
*out_bh = first_bh;
return 0;
failed_bh:
brelse(first_bh);
failed:
return err;
}
/**
* nilfs_mdt_get_block - read or create a buffer on meta data file.
* @inode: inode of the meta data file
* @blkoff: block offset
* @create: create flag
* @init_block: initializer used for newly allocated block
* @out_bh: output of a pointer to the buffer_head
*
* nilfs_mdt_get_block() looks up the specified buffer and tries to create
* a new buffer if @create is not zero. On success, the returned buffer is
* assured to be either existing or formatted using a buffer lock on success.
* @out_bh is substituted only when zero is returned.
*
* Return Value: On success, it returns 0. On error, the following negative
* error code is returned.
*
* %-ENOMEM - Insufficient memory available.
*
* %-EIO - I/O error
*
* %-ENOENT - the specified block does not exist (hole block)
*
* %-EROFS - Read only filesystem (for create mode)
*/
int nilfs_mdt_get_block(struct inode *inode, unsigned long blkoff, int create,
void (*init_block)(struct inode *,
struct buffer_head *, void *),
struct buffer_head **out_bh)
{
int ret;
/* Should be rewritten with merging nilfs_mdt_read_block() */
retry:
ret = nilfs_mdt_read_block(inode, blkoff, !create, out_bh);
if (!create || ret != -ENOENT)
return ret;
ret = nilfs_mdt_create_block(inode, blkoff, out_bh, init_block);
if (unlikely(ret == -EEXIST)) {
/* create = 0; */ /* limit read-create loop retries */
goto retry;
}
return ret;
}
/**
* nilfs_mdt_find_block - find and get a buffer on meta data file.
* @inode: inode of the meta data file
* @start: start block offset (inclusive)
* @end: end block offset (inclusive)
* @blkoff: block offset
* @out_bh: place to store a pointer to buffer_head struct
*
* nilfs_mdt_find_block() looks up an existing block in range of
* [@start, @end] and stores pointer to a buffer head of the block to
* @out_bh, and block offset to @blkoff, respectively. @out_bh and
* @blkoff are substituted only when zero is returned.
*
* Return Value: On success, it returns 0. On error, the following negative
* error code is returned.
*
* %-ENOMEM - Insufficient memory available.
*
* %-EIO - I/O error
*
* %-ENOENT - no block was found in the range
*/
int nilfs_mdt_find_block(struct inode *inode, unsigned long start,
unsigned long end, unsigned long *blkoff,
struct buffer_head **out_bh)
{
__u64 next;
int ret;
if (unlikely(start > end))
return -ENOENT;
ret = nilfs_mdt_read_block(inode, start, true, out_bh);
if (!ret) {
*blkoff = start;
goto out;
}
if (unlikely(ret != -ENOENT || start == ULONG_MAX))
goto out;
ret = nilfs_bmap_seek_key(NILFS_I(inode)->i_bmap, start + 1, &next);
if (!ret) {
if (next <= end) {
ret = nilfs_mdt_read_block(inode, next, true, out_bh);
if (!ret)
*blkoff = next;
} else {
ret = -ENOENT;
}
}
out:
return ret;
}
/**
* nilfs_mdt_delete_block - make a hole on the meta data file.
* @inode: inode of the meta data file
* @block: block offset
*
* Return Value: On success, zero is returned.
* On error, one of the following negative error code is returned.
*
* %-ENOMEM - Insufficient memory available.
*
* %-EIO - I/O error
*/
int nilfs_mdt_delete_block(struct inode *inode, unsigned long block)
{
struct nilfs_inode_info *ii = NILFS_I(inode);
int err;
err = nilfs_bmap_delete(ii->i_bmap, block);
if (!err || err == -ENOENT) {
nilfs_mdt_mark_dirty(inode);
nilfs_mdt_forget_block(inode, block);
}
return err;
}
/**
* nilfs_mdt_forget_block - discard dirty state and try to remove the page
* @inode: inode of the meta data file
* @block: block offset
*
* nilfs_mdt_forget_block() clears a dirty flag of the specified buffer, and
* tries to release the page including the buffer from a page cache.
*
* Return Value: On success, 0 is returned. On error, one of the following
* negative error code is returned.
*
* %-EBUSY - page has an active buffer.
*
* %-ENOENT - page cache has no page addressed by the offset.
*/
int nilfs_mdt_forget_block(struct inode *inode, unsigned long block)
{
pgoff_t index = (pgoff_t)block >>
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
(PAGE_SHIFT - inode->i_blkbits);
struct page *page;
unsigned long first_block;
int ret = 0;
int still_dirty;
page = find_lock_page(inode->i_mapping, index);
if (!page)
return -ENOENT;
wait_on_page_writeback(page);
first_block = (unsigned long)index <<
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
(PAGE_SHIFT - inode->i_blkbits);
if (page_has_buffers(page)) {
struct buffer_head *bh;
bh = nilfs_page_get_nth_block(page, block - first_block);
nilfs_forget_buffer(bh);
}
still_dirty = PageDirty(page);
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
if (still_dirty ||
invalidate_inode_pages2_range(inode->i_mapping, index, index) != 0)
ret = -EBUSY;
return ret;
}
int nilfs_mdt_fetch_dirty(struct inode *inode)
{
struct nilfs_inode_info *ii = NILFS_I(inode);
if (nilfs_bmap_test_and_clear_dirty(ii->i_bmap)) {
set_bit(NILFS_I_DIRTY, &ii->i_state);
return 1;
}
return test_bit(NILFS_I_DIRTY, &ii->i_state);
}
static int
nilfs_mdt_write_page(struct page *page, struct writeback_control *wbc)
{
nilfs2: fix issue with flush kernel thread after remount in RO mode because of driver's internal error or metadata corruption The NILFS2 driver remounts itself in RO mode in the case of discovering metadata corruption (for example, discovering a broken bmap). But usually, this takes place when there have been file system operations before remounting in RO mode. Thereby, NILFS2 driver can be in RO mode with presence of dirty pages in modified inodes' address spaces. It results in flush kernel thread's infinite trying to flush dirty pages in RO mode. As a result, it is possible to see such side effects as: (1) flush kernel thread occupies 50% - 99% of CPU time; (2) system can't be shutdowned without manual power switch off. SYMPTOMS: (1) System log contains error message: "Remounting filesystem read-only". (2) The flush kernel thread occupies 50% - 99% of CPU time. (3) The system can't be shutdowned without manual power switch off. REPRODUCTION PATH: (1) Create volume group with name "unencrypted" by means of vgcreate utility. (2) Run script (prepared by Anthony Doggett <Anthony2486@interfaces.org.uk>): ----------------[BEGIN SCRIPT]-------------------- #!/bin/bash VG=unencrypted #apt-get install nilfs-tools darcs lvcreate --size 2G --name ntest $VG mkfs.nilfs2 -b 1024 -B 8192 /dev/mapper/$VG-ntest mkdir /var/tmp/n mkdir /var/tmp/n/ntest mount /dev/mapper/$VG-ntest /var/tmp/n/ntest mkdir /var/tmp/n/ntest/thedir cd /var/tmp/n/ntest/thedir sleep 2 date darcs init sleep 2 dmesg|tail -n 5 date darcs whatsnew || true date sleep 2 dmesg|tail -n 5 ----------------[END SCRIPT]-------------------- (3) Try to shutdown the system. REPRODUCIBILITY: 100% FIX: This patch implements checking mount state of NILFS2 driver in nilfs_writepage(), nilfs_writepages() and nilfs_mdt_write_page() methods. If it is detected the RO mount state then all dirty pages are simply discarded with warning messages is written in system log. [akpm@linux-foundation.org: fix printk warning] Signed-off-by: Vyacheslav Dubeyko <slava@dubeyko.com> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Cc: Anthony Doggett <Anthony2486@interfaces.org.uk> Cc: ARAI Shun-ichi <hermes@ceres.dti.ne.jp> Cc: Piotr Szymaniak <szarpaj@grubelek.pl> Cc: Zahid Chowdhury <zahid.chowdhury@starsolutions.com> Cc: Elmer Zhang <freeboy6716@gmail.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-30 22:27:48 +00:00
struct inode *inode = page->mapping->host;
struct super_block *sb;
int err = 0;
if (inode && sb_rdonly(inode->i_sb)) {
nilfs2: fix issue with flush kernel thread after remount in RO mode because of driver's internal error or metadata corruption The NILFS2 driver remounts itself in RO mode in the case of discovering metadata corruption (for example, discovering a broken bmap). But usually, this takes place when there have been file system operations before remounting in RO mode. Thereby, NILFS2 driver can be in RO mode with presence of dirty pages in modified inodes' address spaces. It results in flush kernel thread's infinite trying to flush dirty pages in RO mode. As a result, it is possible to see such side effects as: (1) flush kernel thread occupies 50% - 99% of CPU time; (2) system can't be shutdowned without manual power switch off. SYMPTOMS: (1) System log contains error message: "Remounting filesystem read-only". (2) The flush kernel thread occupies 50% - 99% of CPU time. (3) The system can't be shutdowned without manual power switch off. REPRODUCTION PATH: (1) Create volume group with name "unencrypted" by means of vgcreate utility. (2) Run script (prepared by Anthony Doggett <Anthony2486@interfaces.org.uk>): ----------------[BEGIN SCRIPT]-------------------- #!/bin/bash VG=unencrypted #apt-get install nilfs-tools darcs lvcreate --size 2G --name ntest $VG mkfs.nilfs2 -b 1024 -B 8192 /dev/mapper/$VG-ntest mkdir /var/tmp/n mkdir /var/tmp/n/ntest mount /dev/mapper/$VG-ntest /var/tmp/n/ntest mkdir /var/tmp/n/ntest/thedir cd /var/tmp/n/ntest/thedir sleep 2 date darcs init sleep 2 dmesg|tail -n 5 date darcs whatsnew || true date sleep 2 dmesg|tail -n 5 ----------------[END SCRIPT]-------------------- (3) Try to shutdown the system. REPRODUCIBILITY: 100% FIX: This patch implements checking mount state of NILFS2 driver in nilfs_writepage(), nilfs_writepages() and nilfs_mdt_write_page() methods. If it is detected the RO mount state then all dirty pages are simply discarded with warning messages is written in system log. [akpm@linux-foundation.org: fix printk warning] Signed-off-by: Vyacheslav Dubeyko <slava@dubeyko.com> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Cc: Anthony Doggett <Anthony2486@interfaces.org.uk> Cc: ARAI Shun-ichi <hermes@ceres.dti.ne.jp> Cc: Piotr Szymaniak <szarpaj@grubelek.pl> Cc: Zahid Chowdhury <zahid.chowdhury@starsolutions.com> Cc: Elmer Zhang <freeboy6716@gmail.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-30 22:27:48 +00:00
/*
* It means that filesystem was remounted in read-only
* mode because of error or metadata corruption. But we
* have dirty pages that try to be flushed in background.
* So, here we simply discard this dirty page.
*/
nilfs_clear_dirty_page(page, false);
unlock_page(page);
return -EROFS;
}
redirty_page_for_writepage(wbc, page);
unlock_page(page);
if (!inode)
return 0;
sb = inode->i_sb;
if (wbc->sync_mode == WB_SYNC_ALL)
err = nilfs_construct_segment(sb);
else if (wbc->for_reclaim)
nilfs_flush_segment(sb, inode->i_ino);
return err;
}
static const struct address_space_operations def_mdt_aops = {
.dirty_folio = block_dirty_folio,
.invalidate_folio = block_invalidate_folio,
.writepage = nilfs_mdt_write_page,
};
static const struct inode_operations def_mdt_iops;
static const struct file_operations def_mdt_fops;
int nilfs_mdt_init(struct inode *inode, gfp_t gfp_mask, size_t objsz)
{
struct nilfs_mdt_info *mi;
mi = kzalloc(max(sizeof(*mi), objsz), GFP_NOFS);
if (!mi)
return -ENOMEM;
init_rwsem(&mi->mi_sem);
inode->i_private = mi;
inode->i_mode = S_IFREG;
mapping_set_gfp_mask(inode->i_mapping, gfp_mask);
inode->i_op = &def_mdt_iops;
inode->i_fop = &def_mdt_fops;
inode->i_mapping->a_ops = &def_mdt_aops;
return 0;
}
/**
* nilfs_mdt_clear - do cleanup for the metadata file
* @inode: inode of the metadata file
*/
void nilfs_mdt_clear(struct inode *inode)
{
struct nilfs_mdt_info *mdi = NILFS_MDT(inode);
if (mdi->mi_palloc_cache)
nilfs_palloc_destroy_cache(inode);
}
/**
* nilfs_mdt_destroy - release resources used by the metadata file
* @inode: inode of the metadata file
*/
void nilfs_mdt_destroy(struct inode *inode)
{
struct nilfs_mdt_info *mdi = NILFS_MDT(inode);
kfree(mdi->mi_bgl); /* kfree(NULL) is safe */
kfree(mdi);
}
void nilfs_mdt_set_entry_size(struct inode *inode, unsigned int entry_size,
unsigned int header_size)
{
struct nilfs_mdt_info *mi = NILFS_MDT(inode);
mi->mi_entry_size = entry_size;
mi->mi_entries_per_block = i_blocksize(inode) / entry_size;
mi->mi_first_entry_offset = DIV_ROUND_UP(header_size, entry_size);
}
/**
* nilfs_mdt_setup_shadow_map - setup shadow map and bind it to metadata file
* @inode: inode of the metadata file
* @shadow: shadow mapping
*/
int nilfs_mdt_setup_shadow_map(struct inode *inode,
struct nilfs_shadow_map *shadow)
{
struct nilfs_mdt_info *mi = NILFS_MDT(inode);
INIT_LIST_HEAD(&shadow->frozen_buffers);
mm: prevent concurrent unmap_mapping_range() on the same inode Michael Leun reported that running parallel opens on a fuse filesystem can trigger a "kernel BUG at mm/truncate.c:475" Gurudas Pai reported the same bug on NFS. The reason is, unmap_mapping_range() is not prepared for more than one concurrent invocation per inode. For example: thread1: going through a big range, stops in the middle of a vma and stores the restart address in vm_truncate_count. thread2: comes in with a small (e.g. single page) unmap request on the same vma, somewhere before restart_address, finds that the vma was already unmapped up to the restart address and happily returns without doing anything. Another scenario would be two big unmap requests, both having to restart the unmapping and each one setting vm_truncate_count to its own value. This could go on forever without any of them being able to finish. Truncate and hole punching already serialize with i_mutex. Other callers of unmap_mapping_range() do not, and it's difficult to get i_mutex protection for all callers. In particular ->d_revalidate(), which calls invalidate_inode_pages2_range() in fuse, may be called with or without i_mutex. This patch adds a new mutex to 'struct address_space' to prevent running multiple concurrent unmap_mapping_range() on the same mapping. [ We'll hopefully get rid of all this with the upcoming mm preemptibility series by Peter Zijlstra, the "mm: Remove i_mmap_mutex lockbreak" patch in particular. But that is for 2.6.39 ] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Reported-by: Michael Leun <lkml20101129@newton.leun.net> Reported-by: Gurudas Pai <gurudas.pai@oracle.com> Tested-by: Gurudas Pai <gurudas.pai@oracle.com> Acked-by: Hugh Dickins <hughd@google.com> Cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-02-23 12:49:47 +00:00
address_space_init_once(&shadow->frozen_data);
nilfs_mapping_init(&shadow->frozen_data, inode);
mm: prevent concurrent unmap_mapping_range() on the same inode Michael Leun reported that running parallel opens on a fuse filesystem can trigger a "kernel BUG at mm/truncate.c:475" Gurudas Pai reported the same bug on NFS. The reason is, unmap_mapping_range() is not prepared for more than one concurrent invocation per inode. For example: thread1: going through a big range, stops in the middle of a vma and stores the restart address in vm_truncate_count. thread2: comes in with a small (e.g. single page) unmap request on the same vma, somewhere before restart_address, finds that the vma was already unmapped up to the restart address and happily returns without doing anything. Another scenario would be two big unmap requests, both having to restart the unmapping and each one setting vm_truncate_count to its own value. This could go on forever without any of them being able to finish. Truncate and hole punching already serialize with i_mutex. Other callers of unmap_mapping_range() do not, and it's difficult to get i_mutex protection for all callers. In particular ->d_revalidate(), which calls invalidate_inode_pages2_range() in fuse, may be called with or without i_mutex. This patch adds a new mutex to 'struct address_space' to prevent running multiple concurrent unmap_mapping_range() on the same mapping. [ We'll hopefully get rid of all this with the upcoming mm preemptibility series by Peter Zijlstra, the "mm: Remove i_mmap_mutex lockbreak" patch in particular. But that is for 2.6.39 ] Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Reported-by: Michael Leun <lkml20101129@newton.leun.net> Reported-by: Gurudas Pai <gurudas.pai@oracle.com> Tested-by: Gurudas Pai <gurudas.pai@oracle.com> Acked-by: Hugh Dickins <hughd@google.com> Cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-02-23 12:49:47 +00:00
address_space_init_once(&shadow->frozen_btnodes);
nilfs_mapping_init(&shadow->frozen_btnodes, inode);
mi->mi_shadow = shadow;
return 0;
}
/**
* nilfs_mdt_save_to_shadow_map - copy bmap and dirty pages to shadow map
* @inode: inode of the metadata file
*/
int nilfs_mdt_save_to_shadow_map(struct inode *inode)
{
struct nilfs_mdt_info *mi = NILFS_MDT(inode);
struct nilfs_inode_info *ii = NILFS_I(inode);
struct nilfs_shadow_map *shadow = mi->mi_shadow;
int ret;
ret = nilfs_copy_dirty_pages(&shadow->frozen_data, inode->i_mapping);
if (ret)
goto out;
ret = nilfs_copy_dirty_pages(&shadow->frozen_btnodes,
&ii->i_btnode_cache);
if (ret)
goto out;
nilfs_bmap_save(ii->i_bmap, &shadow->bmap_store);
out:
return ret;
}
int nilfs_mdt_freeze_buffer(struct inode *inode, struct buffer_head *bh)
{
struct nilfs_shadow_map *shadow = NILFS_MDT(inode)->mi_shadow;
struct buffer_head *bh_frozen;
struct page *page;
int blkbits = inode->i_blkbits;
page = grab_cache_page(&shadow->frozen_data, bh->b_page->index);
if (!page)
return -ENOMEM;
if (!page_has_buffers(page))
create_empty_buffers(page, 1 << blkbits, 0);
bh_frozen = nilfs_page_get_nth_block(page, bh_offset(bh) >> blkbits);
if (!buffer_uptodate(bh_frozen))
nilfs_copy_buffer(bh_frozen, bh);
if (list_empty(&bh_frozen->b_assoc_buffers)) {
list_add_tail(&bh_frozen->b_assoc_buffers,
&shadow->frozen_buffers);
set_buffer_nilfs_redirected(bh);
} else {
brelse(bh_frozen); /* already frozen */
}
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
return 0;
}
struct buffer_head *
nilfs_mdt_get_frozen_buffer(struct inode *inode, struct buffer_head *bh)
{
struct nilfs_shadow_map *shadow = NILFS_MDT(inode)->mi_shadow;
struct buffer_head *bh_frozen = NULL;
struct page *page;
int n;
page = find_lock_page(&shadow->frozen_data, bh->b_page->index);
if (page) {
if (page_has_buffers(page)) {
n = bh_offset(bh) >> inode->i_blkbits;
bh_frozen = nilfs_page_get_nth_block(page, n);
}
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
}
return bh_frozen;
}
static void nilfs_release_frozen_buffers(struct nilfs_shadow_map *shadow)
{
struct list_head *head = &shadow->frozen_buffers;
struct buffer_head *bh;
while (!list_empty(head)) {
bh = list_first_entry(head, struct buffer_head,
b_assoc_buffers);
list_del_init(&bh->b_assoc_buffers);
brelse(bh); /* drop ref-count to make it releasable */
}
}
/**
* nilfs_mdt_restore_from_shadow_map - restore dirty pages and bmap state
* @inode: inode of the metadata file
*/
void nilfs_mdt_restore_from_shadow_map(struct inode *inode)
{
struct nilfs_mdt_info *mi = NILFS_MDT(inode);
struct nilfs_inode_info *ii = NILFS_I(inode);
struct nilfs_shadow_map *shadow = mi->mi_shadow;
down_write(&mi->mi_sem);
if (mi->mi_palloc_cache)
nilfs_palloc_clear_cache(inode);
nilfs2: fix issue with flush kernel thread after remount in RO mode because of driver's internal error or metadata corruption The NILFS2 driver remounts itself in RO mode in the case of discovering metadata corruption (for example, discovering a broken bmap). But usually, this takes place when there have been file system operations before remounting in RO mode. Thereby, NILFS2 driver can be in RO mode with presence of dirty pages in modified inodes' address spaces. It results in flush kernel thread's infinite trying to flush dirty pages in RO mode. As a result, it is possible to see such side effects as: (1) flush kernel thread occupies 50% - 99% of CPU time; (2) system can't be shutdowned without manual power switch off. SYMPTOMS: (1) System log contains error message: "Remounting filesystem read-only". (2) The flush kernel thread occupies 50% - 99% of CPU time. (3) The system can't be shutdowned without manual power switch off. REPRODUCTION PATH: (1) Create volume group with name "unencrypted" by means of vgcreate utility. (2) Run script (prepared by Anthony Doggett <Anthony2486@interfaces.org.uk>): ----------------[BEGIN SCRIPT]-------------------- #!/bin/bash VG=unencrypted #apt-get install nilfs-tools darcs lvcreate --size 2G --name ntest $VG mkfs.nilfs2 -b 1024 -B 8192 /dev/mapper/$VG-ntest mkdir /var/tmp/n mkdir /var/tmp/n/ntest mount /dev/mapper/$VG-ntest /var/tmp/n/ntest mkdir /var/tmp/n/ntest/thedir cd /var/tmp/n/ntest/thedir sleep 2 date darcs init sleep 2 dmesg|tail -n 5 date darcs whatsnew || true date sleep 2 dmesg|tail -n 5 ----------------[END SCRIPT]-------------------- (3) Try to shutdown the system. REPRODUCIBILITY: 100% FIX: This patch implements checking mount state of NILFS2 driver in nilfs_writepage(), nilfs_writepages() and nilfs_mdt_write_page() methods. If it is detected the RO mount state then all dirty pages are simply discarded with warning messages is written in system log. [akpm@linux-foundation.org: fix printk warning] Signed-off-by: Vyacheslav Dubeyko <slava@dubeyko.com> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Cc: Anthony Doggett <Anthony2486@interfaces.org.uk> Cc: ARAI Shun-ichi <hermes@ceres.dti.ne.jp> Cc: Piotr Szymaniak <szarpaj@grubelek.pl> Cc: Zahid Chowdhury <zahid.chowdhury@starsolutions.com> Cc: Elmer Zhang <freeboy6716@gmail.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-30 22:27:48 +00:00
nilfs_clear_dirty_pages(inode->i_mapping, true);
nilfs_copy_back_pages(inode->i_mapping, &shadow->frozen_data);
nilfs2: fix issue with flush kernel thread after remount in RO mode because of driver's internal error or metadata corruption The NILFS2 driver remounts itself in RO mode in the case of discovering metadata corruption (for example, discovering a broken bmap). But usually, this takes place when there have been file system operations before remounting in RO mode. Thereby, NILFS2 driver can be in RO mode with presence of dirty pages in modified inodes' address spaces. It results in flush kernel thread's infinite trying to flush dirty pages in RO mode. As a result, it is possible to see such side effects as: (1) flush kernel thread occupies 50% - 99% of CPU time; (2) system can't be shutdowned without manual power switch off. SYMPTOMS: (1) System log contains error message: "Remounting filesystem read-only". (2) The flush kernel thread occupies 50% - 99% of CPU time. (3) The system can't be shutdowned without manual power switch off. REPRODUCTION PATH: (1) Create volume group with name "unencrypted" by means of vgcreate utility. (2) Run script (prepared by Anthony Doggett <Anthony2486@interfaces.org.uk>): ----------------[BEGIN SCRIPT]-------------------- #!/bin/bash VG=unencrypted #apt-get install nilfs-tools darcs lvcreate --size 2G --name ntest $VG mkfs.nilfs2 -b 1024 -B 8192 /dev/mapper/$VG-ntest mkdir /var/tmp/n mkdir /var/tmp/n/ntest mount /dev/mapper/$VG-ntest /var/tmp/n/ntest mkdir /var/tmp/n/ntest/thedir cd /var/tmp/n/ntest/thedir sleep 2 date darcs init sleep 2 dmesg|tail -n 5 date darcs whatsnew || true date sleep 2 dmesg|tail -n 5 ----------------[END SCRIPT]-------------------- (3) Try to shutdown the system. REPRODUCIBILITY: 100% FIX: This patch implements checking mount state of NILFS2 driver in nilfs_writepage(), nilfs_writepages() and nilfs_mdt_write_page() methods. If it is detected the RO mount state then all dirty pages are simply discarded with warning messages is written in system log. [akpm@linux-foundation.org: fix printk warning] Signed-off-by: Vyacheslav Dubeyko <slava@dubeyko.com> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Cc: Anthony Doggett <Anthony2486@interfaces.org.uk> Cc: ARAI Shun-ichi <hermes@ceres.dti.ne.jp> Cc: Piotr Szymaniak <szarpaj@grubelek.pl> Cc: Zahid Chowdhury <zahid.chowdhury@starsolutions.com> Cc: Elmer Zhang <freeboy6716@gmail.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-30 22:27:48 +00:00
nilfs_clear_dirty_pages(&ii->i_btnode_cache, true);
nilfs_copy_back_pages(&ii->i_btnode_cache, &shadow->frozen_btnodes);
nilfs_bmap_restore(ii->i_bmap, &shadow->bmap_store);
up_write(&mi->mi_sem);
}
/**
* nilfs_mdt_clear_shadow_map - truncate pages in shadow map caches
* @inode: inode of the metadata file
*/
void nilfs_mdt_clear_shadow_map(struct inode *inode)
{
struct nilfs_mdt_info *mi = NILFS_MDT(inode);
struct nilfs_shadow_map *shadow = mi->mi_shadow;
down_write(&mi->mi_sem);
nilfs_release_frozen_buffers(shadow);
truncate_inode_pages(&shadow->frozen_data, 0);
truncate_inode_pages(&shadow->frozen_btnodes, 0);
up_write(&mi->mi_sem);
}