linux/fs/f2fs/file.c

2832 lines
66 KiB
C
Raw Normal View History

/*
* fs/f2fs/file.c
*
* Copyright (c) 2012 Samsung Electronics Co., Ltd.
* http://www.samsung.com/
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/fs.h>
#include <linux/f2fs_fs.h>
#include <linux/stat.h>
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/blkdev.h>
#include <linux/falloc.h>
#include <linux/types.h>
#include <linux/compat.h>
#include <linux/uaccess.h>
#include <linux/mount.h>
#include <linux/pagevec.h>
#include <linux/uio.h>
#include <linux/uuid.h>
#include <linux/file.h>
#include "f2fs.h"
#include "node.h"
#include "segment.h"
#include "xattr.h"
#include "acl.h"
#include "gc.h"
#include "trace.h"
#include <trace/events/f2fs.h>
static int f2fs_filemap_fault(struct vm_fault *vmf)
{
struct inode *inode = file_inode(vmf->vma->vm_file);
int err;
down_read(&F2FS_I(inode)->i_mmap_sem);
err = filemap_fault(vmf);
up_read(&F2FS_I(inode)->i_mmap_sem);
return err;
}
static int f2fs_vm_page_mkwrite(struct vm_fault *vmf)
{
struct page *page = vmf->page;
struct inode *inode = file_inode(vmf->vma->vm_file);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct dnode_of_data dn;
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
int err;
if (unlikely(f2fs_cp_error(sbi))) {
err = -EIO;
goto err;
}
sb_start_pagefault(inode->i_sb);
f2fs_bug_on(sbi, f2fs_has_inline_data(inode));
/* block allocation */
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
f2fs_lock_op(sbi);
set_new_dnode(&dn, inode, NULL, NULL, 0);
err = f2fs_reserve_block(&dn, page->index);
if (err) {
f2fs_unlock_op(sbi);
goto out;
}
f2fs_put_dnode(&dn);
f2fs_unlock_op(sbi);
f2fs_balance_fs(sbi, dn.node_changed);
file_update_time(vmf->vma->vm_file);
down_read(&F2FS_I(inode)->i_mmap_sem);
lock_page(page);
if (unlikely(page->mapping != inode->i_mapping ||
page_offset(page) > i_size_read(inode) ||
!PageUptodate(page))) {
unlock_page(page);
err = -EFAULT;
goto out_sem;
}
/*
* check to see if the page is mapped already (no holes)
*/
if (PageMappedToDisk(page))
goto mapped;
/* page is wholly or partially inside EOF */
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
if (((loff_t)(page->index + 1) << PAGE_SHIFT) >
i_size_read(inode)) {
unsigned offset;
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
offset = i_size_read(inode) & ~PAGE_MASK;
zero_user_segment(page, offset, PAGE_SIZE);
}
set_page_dirty(page);
if (!PageUptodate(page))
SetPageUptodate(page);
f2fs_update_iostat(sbi, APP_MAPPED_IO, F2FS_BLKSIZE);
trace_f2fs_vm_page_mkwrite(page, DATA);
mapped:
/* fill the page */
f2fs_wait_on_page_writeback(page, DATA, false);
f2fs crypto: fix racing of accessing encrypted page among different competitors Since we use different page cache (normally inode's page cache for R/W and meta inode's page cache for GC) to cache the same physical block which is belong to an encrypted inode. Writeback of these two page cache should be exclusive, but now we didn't handle writeback state well, so there may be potential racing problem: a) kworker: f2fs_gc: - f2fs_write_data_pages - f2fs_write_data_page - do_write_data_page - write_data_page - f2fs_submit_page_mbio (page#1 in inode's page cache was queued in f2fs bio cache, and be ready to write to new blkaddr) - gc_data_segment - move_encrypted_block - pagecache_get_page (page#2 in meta inode's page cache was cached with the invalid datas of physical block located in new blkaddr) - f2fs_submit_page_mbio (page#1 was submitted, later, page#2 with invalid data will be submitted) b) f2fs_gc: - gc_data_segment - move_encrypted_block - f2fs_submit_page_mbio (page#1 in meta inode's page cache was queued in f2fs bio cache, and be ready to write to new blkaddr) user thread: - f2fs_write_begin - f2fs_submit_page_bio (we submit the request to block layer to update page#2 in inode's page cache with physical block located in new blkaddr, so here we may read gabbage data from new blkaddr since GC hasn't writebacked the page#1 yet) This patch fixes above potential racing problem for encrypted inode. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-10-08 05:27:34 +00:00
/* wait for GCed encrypted page writeback */
if (f2fs_encrypted_file(inode))
f2fs_wait_on_block_writeback(sbi, dn.data_blkaddr);
f2fs crypto: fix racing of accessing encrypted page among different competitors Since we use different page cache (normally inode's page cache for R/W and meta inode's page cache for GC) to cache the same physical block which is belong to an encrypted inode. Writeback of these two page cache should be exclusive, but now we didn't handle writeback state well, so there may be potential racing problem: a) kworker: f2fs_gc: - f2fs_write_data_pages - f2fs_write_data_page - do_write_data_page - write_data_page - f2fs_submit_page_mbio (page#1 in inode's page cache was queued in f2fs bio cache, and be ready to write to new blkaddr) - gc_data_segment - move_encrypted_block - pagecache_get_page (page#2 in meta inode's page cache was cached with the invalid datas of physical block located in new blkaddr) - f2fs_submit_page_mbio (page#1 was submitted, later, page#2 with invalid data will be submitted) b) f2fs_gc: - gc_data_segment - move_encrypted_block - f2fs_submit_page_mbio (page#1 in meta inode's page cache was queued in f2fs bio cache, and be ready to write to new blkaddr) user thread: - f2fs_write_begin - f2fs_submit_page_bio (we submit the request to block layer to update page#2 in inode's page cache with physical block located in new blkaddr, so here we may read gabbage data from new blkaddr since GC hasn't writebacked the page#1 yet) This patch fixes above potential racing problem for encrypted inode. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-10-08 05:27:34 +00:00
out_sem:
up_read(&F2FS_I(inode)->i_mmap_sem);
out:
sb_end_pagefault(inode->i_sb);
f2fs_update_time(sbi, REQ_TIME);
err:
return block_page_mkwrite_return(err);
}
static const struct vm_operations_struct f2fs_file_vm_ops = {
.fault = f2fs_filemap_fault,
.map_pages = filemap_map_pages,
.page_mkwrite = f2fs_vm_page_mkwrite,
};
static int get_parent_ino(struct inode *inode, nid_t *pino)
{
struct dentry *dentry;
inode = igrab(inode);
dentry = d_find_any_alias(inode);
iput(inode);
if (!dentry)
return 0;
*pino = parent_ino(dentry);
dput(dentry);
return 1;
}
static inline enum cp_reason_type need_do_checkpoint(struct inode *inode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
enum cp_reason_type cp_reason = CP_NO_NEEDED;
if (!S_ISREG(inode->i_mode))
cp_reason = CP_NON_REGULAR;
else if (inode->i_nlink != 1)
cp_reason = CP_HARDLINK;
else if (is_sbi_flag_set(sbi, SBI_NEED_CP))
cp_reason = CP_SB_NEED_CP;
else if (file_wrong_pino(inode))
cp_reason = CP_WRONG_PINO;
else if (!space_for_roll_forward(sbi))
cp_reason = CP_NO_SPC_ROLL;
else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
cp_reason = CP_NODE_NEED_CP;
else if (test_opt(sbi, FASTBOOT))
cp_reason = CP_FASTBOOT_MODE;
else if (sbi->active_logs == 2)
cp_reason = CP_SPEC_LOG_NUM;
return cp_reason;
}
static bool need_inode_page_update(struct f2fs_sb_info *sbi, nid_t ino)
{
struct page *i = find_get_page(NODE_MAPPING(sbi), ino);
bool ret = false;
/* But we need to avoid that there are some inode updates */
if ((i && PageDirty(i)) || need_inode_block_update(sbi, ino))
ret = true;
f2fs_put_page(i, 0);
return ret;
}
static void try_to_fix_pino(struct inode *inode)
{
struct f2fs_inode_info *fi = F2FS_I(inode);
nid_t pino;
down_write(&fi->i_sem);
if (file_wrong_pino(inode) && inode->i_nlink == 1 &&
get_parent_ino(inode, &pino)) {
f2fs_i_pino_write(inode, pino);
file_got_pino(inode);
}
up_write(&fi->i_sem);
}
static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
int datasync, bool atomic)
{
struct inode *inode = file->f_mapping->host;
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
nid_t ino = inode->i_ino;
int ret = 0;
enum cp_reason_type cp_reason = 0;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_ALL,
.nr_to_write = LONG_MAX,
.for_reclaim = 0,
};
if (unlikely(f2fs_readonly(inode->i_sb)))
return 0;
trace_f2fs_sync_file_enter(inode);
/* if fdatasync is triggered, let's do in-place-update */
if (datasync || get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
set_inode_flag(inode, FI_NEED_IPU);
ret = file_write_and_wait_range(file, start, end);
clear_inode_flag(inode, FI_NEED_IPU);
if (ret) {
trace_f2fs_sync_file_exit(inode, cp_reason, datasync, ret);
return ret;
}
/* if the inode is dirty, let's recover all the time */
if (!f2fs_skip_inode_update(inode, datasync)) {
f2fs_write_inode(inode, NULL);
goto go_write;
}
/*
* if there is no written data, don't waste time to write recovery info.
*/
if (!is_inode_flag_set(inode, FI_APPEND_WRITE) &&
!exist_written_data(sbi, ino, APPEND_INO)) {
/* it may call write_inode just prior to fsync */
if (need_inode_page_update(sbi, ino))
goto go_write;
if (is_inode_flag_set(inode, FI_UPDATE_WRITE) ||
exist_written_data(sbi, ino, UPDATE_INO))
goto flush_out;
goto out;
}
go_write:
/*
* Both of fdatasync() and fsync() are able to be recovered from
* sudden-power-off.
*/
down_read(&F2FS_I(inode)->i_sem);
cp_reason = need_do_checkpoint(inode);
up_read(&F2FS_I(inode)->i_sem);
if (cp_reason) {
/* all the dirty node pages should be flushed for POR */
ret = f2fs_sync_fs(inode->i_sb, 1);
/*
* We've secured consistency through sync_fs. Following pino
* will be used only for fsynced inodes after checkpoint.
*/
try_to_fix_pino(inode);
clear_inode_flag(inode, FI_APPEND_WRITE);
clear_inode_flag(inode, FI_UPDATE_WRITE);
goto out;
}
f2fs: fix conditions to remain recovery information in f2fs_sync_file This patch revisited whole the recovery information during the f2fs_sync_file. In this patch, there are three information to make a decision. a) IS_CHECKPOINTED, /* is it checkpointed before? */ b) HAS_FSYNCED_INODE, /* is the inode fsynced before? */ c) HAS_LAST_FSYNC, /* has the latest node fsync mark? */ And, the scenarios for our rule are based on: [Term] F: fsync_mark, D: dentry_mark 1. inode(x) | CP | inode(x) | dnode(F) 2. inode(x) | CP | inode(F) | dnode(F) 3. inode(x) | CP | dnode(F) | inode(x) | inode(F) 4. inode(x) | CP | dnode(F) | inode(F) 5. CP | inode(x) | dnode(F) | inode(DF) 6. CP | inode(DF) | dnode(F) 7. CP | dnode(F) | inode(DF) 8. CP | dnode(F) | inode(x) | inode(DF) For example, #3, the three conditions should be changed as follows. inode(x) | CP | dnode(F) | inode(x) | inode(F) a) x o o o o b) x x x x o c) x o o x o If f2fs_sync_file stops ------^, it should write inode(F) --------------^ So, the need_inode_block_update should return true, since c) get_nat_flag(e, HAS_LAST_FSYNC), is false. For example, #8, CP | alloc | dnode(F) | inode(x) | inode(DF) a) o x x x x b) x x x o c) o o x o If f2fs_sync_file stops -------^, it should write inode(DF) --------------^ Note that, the roll-forward policy should follow this rule, which means, if there are any missing blocks, we doesn't need to recover that inode. Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-09-15 21:50:48 +00:00
sync_nodes:
ret = fsync_node_pages(sbi, inode, &wbc, atomic);
if (ret)
goto out;
/* if cp_error was enabled, we should avoid infinite loop */
if (unlikely(f2fs_cp_error(sbi))) {
ret = -EIO;
goto out;
}
if (need_inode_block_update(sbi, ino)) {
f2fs_mark_inode_dirty_sync(inode, true);
f2fs_write_inode(inode, NULL);
goto sync_nodes;
}
/*
* If it's atomic_write, it's just fine to keep write ordering. So
* here we don't need to wait for node write completion, since we use
* node chain which serializes node blocks. If one of node writes are
* reordered, we can see simply broken chain, resulting in stopping
* roll-forward recovery. It means we'll recover all or none node blocks
* given fsync mark.
*/
if (!atomic) {
ret = wait_on_node_pages_writeback(sbi, ino);
if (ret)
goto out;
}
/* once recovery info is written, don't need to tack this */
remove_ino_entry(sbi, ino, APPEND_INO);
clear_inode_flag(inode, FI_APPEND_WRITE);
flush_out:
if (!atomic)
ret = f2fs_issue_flush(sbi, inode->i_ino);
if (!ret) {
remove_ino_entry(sbi, ino, UPDATE_INO);
clear_inode_flag(inode, FI_UPDATE_WRITE);
remove_ino_entry(sbi, ino, FLUSH_INO);
}
f2fs_update_time(sbi, REQ_TIME);
out:
trace_f2fs_sync_file_exit(inode, cp_reason, datasync, ret);
f2fs_trace_ios(NULL, 1);
return ret;
}
int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
if (unlikely(f2fs_cp_error(F2FS_I_SB(file_inode(file)))))
return -EIO;
return f2fs_do_sync_file(file, start, end, datasync, false);
}
static pgoff_t __get_first_dirty_index(struct address_space *mapping,
pgoff_t pgofs, int whence)
{
struct page *page;
int nr_pages;
if (whence != SEEK_DATA)
return 0;
/* find first dirty page index */
nr_pages = find_get_pages_tag(mapping, &pgofs, PAGECACHE_TAG_DIRTY,
1, &page);
if (!nr_pages)
return ULONG_MAX;
pgofs = page->index;
put_page(page);
return pgofs;
}
static bool __found_offset(block_t blkaddr, pgoff_t dirty, pgoff_t pgofs,
int whence)
{
switch (whence) {
case SEEK_DATA:
if ((blkaddr == NEW_ADDR && dirty == pgofs) ||
(blkaddr != NEW_ADDR && blkaddr != NULL_ADDR))
return true;
break;
case SEEK_HOLE:
if (blkaddr == NULL_ADDR)
return true;
break;
}
return false;
}
static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
{
struct inode *inode = file->f_mapping->host;
loff_t maxbytes = inode->i_sb->s_maxbytes;
struct dnode_of_data dn;
pgoff_t pgofs, end_offset, dirty;
loff_t data_ofs = offset;
loff_t isize;
int err = 0;
inode_lock(inode);
isize = i_size_read(inode);
if (offset >= isize)
goto fail;
/* handle inline data case */
if (f2fs_has_inline_data(inode) || f2fs_has_inline_dentry(inode)) {
if (whence == SEEK_HOLE)
data_ofs = isize;
goto found;
}
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
pgofs = (pgoff_t)(offset >> PAGE_SHIFT);
dirty = __get_first_dirty_index(inode->i_mapping, pgofs, whence);
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
for (; data_ofs < isize; data_ofs = (loff_t)pgofs << PAGE_SHIFT) {
set_new_dnode(&dn, inode, NULL, NULL, 0);
err = get_dnode_of_data(&dn, pgofs, LOOKUP_NODE);
if (err && err != -ENOENT) {
goto fail;
} else if (err == -ENOENT) {
/* direct node does not exists */
if (whence == SEEK_DATA) {
pgofs = get_next_page_offset(&dn, pgofs);
continue;
} else {
goto found;
}
}
end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
/* find data/hole in dnode block */
for (; dn.ofs_in_node < end_offset;
dn.ofs_in_node++, pgofs++,
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
data_ofs = (loff_t)pgofs << PAGE_SHIFT) {
block_t blkaddr;
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
blkaddr = datablock_addr(dn.inode,
dn.node_page, dn.ofs_in_node);
if (__found_offset(blkaddr, dirty, pgofs, whence)) {
f2fs_put_dnode(&dn);
goto found;
}
}
f2fs_put_dnode(&dn);
}
if (whence == SEEK_DATA)
goto fail;
found:
if (whence == SEEK_HOLE && data_ofs > isize)
data_ofs = isize;
inode_unlock(inode);
return vfs_setpos(file, data_ofs, maxbytes);
fail:
inode_unlock(inode);
return -ENXIO;
}
static loff_t f2fs_llseek(struct file *file, loff_t offset, int whence)
{
struct inode *inode = file->f_mapping->host;
loff_t maxbytes = inode->i_sb->s_maxbytes;
switch (whence) {
case SEEK_SET:
case SEEK_CUR:
case SEEK_END:
return generic_file_llseek_size(file, offset, whence,
maxbytes, i_size_read(inode));
case SEEK_DATA:
case SEEK_HOLE:
if (offset < 0)
return -ENXIO;
return f2fs_seek_block(file, offset, whence);
}
return -EINVAL;
}
static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
struct inode *inode = file_inode(file);
int err;
if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
return -EIO;
/* we don't need to use inline_data strictly */
err = f2fs_convert_inline_inode(inode);
if (err)
return err;
file_accessed(file);
vma->vm_ops = &f2fs_file_vm_ops;
return 0;
}
static int f2fs_file_open(struct inode *inode, struct file *filp)
{
struct dentry *dir;
if (f2fs_encrypted_inode(inode)) {
int ret = fscrypt_get_encryption_info(inode);
if (ret)
return -EACCES;
if (!fscrypt_has_encryption_key(inode))
return -ENOKEY;
}
dir = dget_parent(file_dentry(filp));
if (f2fs_encrypted_inode(d_inode(dir)) &&
!fscrypt_has_permitted_context(d_inode(dir), inode)) {
dput(dir);
return -EPERM;
}
dput(dir);
return dquot_file_open(inode, filp);
}
f2fs: reuse the locked dnode page and its inode This patch fixes the following deadlock bug during the recovery. INFO: task mount:1322 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. mount D ffffffff81125870 0 1322 1266 0x00000000 ffff8801207e39d8 0000000000000046 ffff88012ab1dee0 0000000000000046 ffff8801207e3a08 ffff880115903f40 ffff8801207e3fd8 ffff8801207e3fd8 ffff8801207e3fd8 ffff880115903f40 ffff8801207e39d8 ffff88012fc94520 Call Trace: [<ffffffff81125870>] ? __lock_page+0x70/0x70 [<ffffffff816a92d9>] schedule+0x29/0x70 [<ffffffff816a93af>] io_schedule+0x8f/0xd0 [<ffffffff8112587e>] sleep_on_page+0xe/0x20 [<ffffffff816a649a>] __wait_on_bit_lock+0x5a/0xc0 [<ffffffff81125867>] __lock_page+0x67/0x70 [<ffffffff8106c7b0>] ? autoremove_wake_function+0x40/0x40 [<ffffffff81126857>] find_lock_page+0x67/0x80 [<ffffffff8112698f>] find_or_create_page+0x3f/0xb0 [<ffffffffa03901a8>] ? sync_inode_page+0xa8/0xd0 [f2fs] [<ffffffffa038fdf7>] get_node_page+0x67/0x180 [f2fs] [<ffffffffa039818b>] recover_fsync_data+0xacb/0xff0 [f2fs] [<ffffffff816aaa1e>] ? _raw_spin_unlock+0x3e/0x40 [<ffffffffa0389634>] f2fs_fill_super+0x7d4/0x850 [f2fs] [<ffffffff81184cf9>] mount_bdev+0x1c9/0x210 [<ffffffffa0388e60>] ? validate_superblock+0x180/0x180 [f2fs] [<ffffffffa0387635>] f2fs_mount+0x15/0x20 [f2fs] [<ffffffff81185a13>] mount_fs+0x43/0x1b0 [<ffffffff81145ba0>] ? __alloc_percpu+0x10/0x20 [<ffffffff811a0796>] vfs_kern_mount+0x76/0x120 [<ffffffff811a2cb7>] do_mount+0x237/0xa10 [<ffffffff81140b9b>] ? strndup_user+0x5b/0x80 [<ffffffff811a3520>] SyS_mount+0x90/0xe0 [<ffffffff816b3502>] system_call_fastpath+0x16/0x1b The bug is triggered when check_index_in_prev_nodes tries to get the direct node page by calling get_node_page. At this point, if the direct node page is already locked by get_dnode_of_data, its caller, we got a deadlock condition. This patch adds additional condition check for the reuse of locked direct node pages prior to the get_node_page call. Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-21 23:02:02 +00:00
int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
struct f2fs_node *raw_node;
f2fs: update extent tree in batches This patch introduce a new helper f2fs_update_extent_tree_range which can do extent mapping update at a specified range. The main idea is: 1) punch all mapping info in extent node(s) which are at a specified range; 2) try to merge new extent mapping with adjacent node, or failing that, insert the mapping into extent tree as a new node. In order to see the benefit, I add a function for stating time stamping count as below: uint64_t rdtsc(void) { uint32_t lo, hi; __asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi)); return (uint64_t)hi << 32 | lo; } My test environment is: ubuntu, intel i7-3770, 16G memory, 256g micron ssd. truncation path: update extent cache from truncate_data_blocks_range non-truncataion path: update extent cache from other paths total: all update paths a) Removing 128MB file which has one extent node mapping whole range of file: 1. dd if=/dev/zero of=/mnt/f2fs/128M bs=1M count=128 2. sync 3. rm /mnt/f2fs/128M Before: total count average truncation: 7651022 32768 233.49 Patched: total count average truncation: 3321 33 100.64 b) fsstress: fsstress -d /mnt/f2fs -l 5 -n 100 -p 20 Test times: 5 times. Before: total count average truncation: 5812480.6 20911.6 277.95 non-truncation: 7783845.6 13440.8 579.12 total: 13596326.2 34352.4 395.79 Patched: total count average truncation: 1281283.0 3041.6 421.25 non-truncation: 7355844.4 13662.8 538.38 total: 8637127.4 16704.4 517.06 1) For the updates in truncation path: - we can see updating in batches leads total tsc and update count reducing explicitly; - besides, for a single batched updating, punching multiple extent nodes in a loop, result in executing more operations, so our average tsc increase intensively. 2) For the updates in non-truncation path: - there is a little improvement, that is because for the scenario that we just need to update in the head or tail of extent node, new interface optimize to update info in extent node directly, rather than removing original extent node for updating and then inserting that updated one into cache as new node. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-08-26 12:34:48 +00:00
int nr_free = 0, ofs = dn->ofs_in_node, len = count;
__le32 *addr;
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
int base = 0;
if (IS_INODE(dn->node_page) && f2fs_has_extra_attr(dn->inode))
base = get_extra_isize(dn->inode);
raw_node = F2FS_NODE(dn->node_page);
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
addr = blkaddr_in_node(raw_node) + base + ofs;
for (; count > 0; count--, addr++, dn->ofs_in_node++) {
block_t blkaddr = le32_to_cpu(*addr);
if (blkaddr == NULL_ADDR)
continue;
dn->data_blkaddr = NULL_ADDR;
set_data_blkaddr(dn);
invalidate_blocks(sbi, blkaddr);
if (dn->ofs_in_node == 0 && IS_INODE(dn->node_page))
clear_inode_flag(dn->inode, FI_FIRST_BLOCK_WRITTEN);
nr_free++;
}
f2fs: update extent tree in batches This patch introduce a new helper f2fs_update_extent_tree_range which can do extent mapping update at a specified range. The main idea is: 1) punch all mapping info in extent node(s) which are at a specified range; 2) try to merge new extent mapping with adjacent node, or failing that, insert the mapping into extent tree as a new node. In order to see the benefit, I add a function for stating time stamping count as below: uint64_t rdtsc(void) { uint32_t lo, hi; __asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi)); return (uint64_t)hi << 32 | lo; } My test environment is: ubuntu, intel i7-3770, 16G memory, 256g micron ssd. truncation path: update extent cache from truncate_data_blocks_range non-truncataion path: update extent cache from other paths total: all update paths a) Removing 128MB file which has one extent node mapping whole range of file: 1. dd if=/dev/zero of=/mnt/f2fs/128M bs=1M count=128 2. sync 3. rm /mnt/f2fs/128M Before: total count average truncation: 7651022 32768 233.49 Patched: total count average truncation: 3321 33 100.64 b) fsstress: fsstress -d /mnt/f2fs -l 5 -n 100 -p 20 Test times: 5 times. Before: total count average truncation: 5812480.6 20911.6 277.95 non-truncation: 7783845.6 13440.8 579.12 total: 13596326.2 34352.4 395.79 Patched: total count average truncation: 1281283.0 3041.6 421.25 non-truncation: 7355844.4 13662.8 538.38 total: 8637127.4 16704.4 517.06 1) For the updates in truncation path: - we can see updating in batches leads total tsc and update count reducing explicitly; - besides, for a single batched updating, punching multiple extent nodes in a loop, result in executing more operations, so our average tsc increase intensively. 2) For the updates in non-truncation path: - there is a little improvement, that is because for the scenario that we just need to update in the head or tail of extent node, new interface optimize to update info in extent node directly, rather than removing original extent node for updating and then inserting that updated one into cache as new node. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-08-26 12:34:48 +00:00
if (nr_free) {
f2fs: update extent tree in batches This patch introduce a new helper f2fs_update_extent_tree_range which can do extent mapping update at a specified range. The main idea is: 1) punch all mapping info in extent node(s) which are at a specified range; 2) try to merge new extent mapping with adjacent node, or failing that, insert the mapping into extent tree as a new node. In order to see the benefit, I add a function for stating time stamping count as below: uint64_t rdtsc(void) { uint32_t lo, hi; __asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi)); return (uint64_t)hi << 32 | lo; } My test environment is: ubuntu, intel i7-3770, 16G memory, 256g micron ssd. truncation path: update extent cache from truncate_data_blocks_range non-truncataion path: update extent cache from other paths total: all update paths a) Removing 128MB file which has one extent node mapping whole range of file: 1. dd if=/dev/zero of=/mnt/f2fs/128M bs=1M count=128 2. sync 3. rm /mnt/f2fs/128M Before: total count average truncation: 7651022 32768 233.49 Patched: total count average truncation: 3321 33 100.64 b) fsstress: fsstress -d /mnt/f2fs -l 5 -n 100 -p 20 Test times: 5 times. Before: total count average truncation: 5812480.6 20911.6 277.95 non-truncation: 7783845.6 13440.8 579.12 total: 13596326.2 34352.4 395.79 Patched: total count average truncation: 1281283.0 3041.6 421.25 non-truncation: 7355844.4 13662.8 538.38 total: 8637127.4 16704.4 517.06 1) For the updates in truncation path: - we can see updating in batches leads total tsc and update count reducing explicitly; - besides, for a single batched updating, punching multiple extent nodes in a loop, result in executing more operations, so our average tsc increase intensively. 2) For the updates in non-truncation path: - there is a little improvement, that is because for the scenario that we just need to update in the head or tail of extent node, new interface optimize to update info in extent node directly, rather than removing original extent node for updating and then inserting that updated one into cache as new node. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-08-26 12:34:48 +00:00
pgoff_t fofs;
/*
* once we invalidate valid blkaddr in range [ofs, ofs + count],
* we will invalidate all blkaddr in the whole range.
*/
fofs = start_bidx_of_node(ofs_of_node(dn->node_page),
dn->inode) + ofs;
f2fs: update extent tree in batches This patch introduce a new helper f2fs_update_extent_tree_range which can do extent mapping update at a specified range. The main idea is: 1) punch all mapping info in extent node(s) which are at a specified range; 2) try to merge new extent mapping with adjacent node, or failing that, insert the mapping into extent tree as a new node. In order to see the benefit, I add a function for stating time stamping count as below: uint64_t rdtsc(void) { uint32_t lo, hi; __asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi)); return (uint64_t)hi << 32 | lo; } My test environment is: ubuntu, intel i7-3770, 16G memory, 256g micron ssd. truncation path: update extent cache from truncate_data_blocks_range non-truncataion path: update extent cache from other paths total: all update paths a) Removing 128MB file which has one extent node mapping whole range of file: 1. dd if=/dev/zero of=/mnt/f2fs/128M bs=1M count=128 2. sync 3. rm /mnt/f2fs/128M Before: total count average truncation: 7651022 32768 233.49 Patched: total count average truncation: 3321 33 100.64 b) fsstress: fsstress -d /mnt/f2fs -l 5 -n 100 -p 20 Test times: 5 times. Before: total count average truncation: 5812480.6 20911.6 277.95 non-truncation: 7783845.6 13440.8 579.12 total: 13596326.2 34352.4 395.79 Patched: total count average truncation: 1281283.0 3041.6 421.25 non-truncation: 7355844.4 13662.8 538.38 total: 8637127.4 16704.4 517.06 1) For the updates in truncation path: - we can see updating in batches leads total tsc and update count reducing explicitly; - besides, for a single batched updating, punching multiple extent nodes in a loop, result in executing more operations, so our average tsc increase intensively. 2) For the updates in non-truncation path: - there is a little improvement, that is because for the scenario that we just need to update in the head or tail of extent node, new interface optimize to update info in extent node directly, rather than removing original extent node for updating and then inserting that updated one into cache as new node. Signed-off-by: Chao Yu <chao2.yu@samsung.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-08-26 12:34:48 +00:00
f2fs_update_extent_cache_range(dn, fofs, 0, len);
dec_valid_block_count(sbi, dn->inode, nr_free);
}
dn->ofs_in_node = ofs;
f2fs_update_time(sbi, REQ_TIME);
trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
dn->ofs_in_node, nr_free);
return nr_free;
}
void truncate_data_blocks(struct dnode_of_data *dn)
{
truncate_data_blocks_range(dn, ADDRS_PER_BLOCK);
}
static int truncate_partial_data_page(struct inode *inode, u64 from,
bool cache_only)
{
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
unsigned offset = from & (PAGE_SIZE - 1);
pgoff_t index = from >> PAGE_SHIFT;
struct address_space *mapping = inode->i_mapping;
struct page *page;
if (!offset && !cache_only)
return 0;
if (cache_only) {
page = find_lock_page(mapping, index);
if (page && PageUptodate(page))
goto truncate_out;
f2fs_put_page(page, 1);
return 0;
}
page = get_lock_data_page(inode, index, true);
if (IS_ERR(page))
return PTR_ERR(page) == -ENOENT ? 0 : PTR_ERR(page);
truncate_out:
f2fs_wait_on_page_writeback(page, DATA, true);
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
zero_user(page, offset, PAGE_SIZE - offset);
/* An encrypted inode should have a key and truncate the last page. */
f2fs_bug_on(F2FS_I_SB(inode), cache_only && f2fs_encrypted_inode(inode));
if (!cache_only)
set_page_dirty(page);
f2fs_put_page(page, 1);
return 0;
}
int truncate_blocks(struct inode *inode, u64 from, bool lock)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
unsigned int blocksize = inode->i_sb->s_blocksize;
struct dnode_of_data dn;
pgoff_t free_from;
int count = 0, err = 0;
struct page *ipage;
bool truncate_page = false;
trace_f2fs_truncate_blocks_enter(inode, from);
free_from = (pgoff_t)F2FS_BYTES_TO_BLK(from + blocksize - 1);
if (free_from >= sbi->max_file_blocks)
goto free_partial;
if (lock)
f2fs_lock_op(sbi);
ipage = get_node_page(sbi, inode->i_ino);
if (IS_ERR(ipage)) {
err = PTR_ERR(ipage);
goto out;
}
if (f2fs_has_inline_data(inode)) {
truncate_inline_inode(inode, ipage, from);
f2fs_put_page(ipage, 1);
truncate_page = true;
goto out;
}
set_new_dnode(&dn, inode, ipage, NULL, 0);
err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE_RA);
if (err) {
if (err == -ENOENT)
goto free_next;
goto out;
}
count = ADDRS_PER_PAGE(dn.node_page, inode);
count -= dn.ofs_in_node;
f2fs_bug_on(sbi, count < 0);
f2fs: introduce a new global lock scheme In the previous version, f2fs uses global locks according to the usage types, such as directory operations, block allocation, block write, and so on. Reference the following lock types in f2fs.h. enum lock_type { RENAME, /* for renaming operations */ DENTRY_OPS, /* for directory operations */ DATA_WRITE, /* for data write */ DATA_NEW, /* for data allocation */ DATA_TRUNC, /* for data truncate */ NODE_NEW, /* for node allocation */ NODE_TRUNC, /* for node truncate */ NODE_WRITE, /* for node write */ NR_LOCK_TYPE, }; In that case, we lose the performance under the multi-threading environment, since every types of operations must be conducted one at a time. In order to address the problem, let's share the locks globally with a mutex array regardless of any types. So, let users grab a mutex and perform their jobs in parallel as much as possbile. For this, I propose a new global lock scheme as follows. 0. Data structure - f2fs_sb_info -> mutex_lock[NR_GLOBAL_LOCKS] - f2fs_sb_info -> node_write 1. mutex_lock_op(sbi) - try to get an avaiable lock from the array. - returns the index of the gottern lock variable. 2. mutex_unlock_op(sbi, index of the lock) - unlock the given index of the lock. 3. mutex_lock_all(sbi) - grab all the locks in the array before the checkpoint. 4. mutex_unlock_all(sbi) - release all the locks in the array after checkpoint. 5. block_operations() - call mutex_lock_all() - sync_dirty_dir_inodes() - grab node_write - sync_node_pages() Note that, the pairs of mutex_lock_op()/mutex_unlock_op() and mutex_lock_all()/mutex_unlock_all() should be used together. Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-22 07:21:29 +00:00
if (dn.ofs_in_node || IS_INODE(dn.node_page)) {
truncate_data_blocks_range(&dn, count);
free_from += count;
}
f2fs_put_dnode(&dn);
free_next:
err = truncate_inode_blocks(inode, free_from);
out:
if (lock)
f2fs_unlock_op(sbi);
free_partial:
/* lastly zero out the first data page */
if (!err)
err = truncate_partial_data_page(inode, from, truncate_page);
trace_f2fs_truncate_blocks_exit(inode, err);
return err;
}
int f2fs_truncate(struct inode *inode)
{
int err;
if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
return -EIO;
if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
S_ISLNK(inode->i_mode)))
return 0;
trace_f2fs_truncate(inode);
#ifdef CONFIG_F2FS_FAULT_INJECTION
if (time_to_inject(F2FS_I_SB(inode), FAULT_TRUNCATE)) {
f2fs_show_injection_info(FAULT_TRUNCATE);
return -EIO;
}
#endif
/* we should check inline_data size */
if (!f2fs_may_inline_data(inode)) {
err = f2fs_convert_inline_inode(inode);
if (err)
return err;
}
err = truncate_blocks(inode, i_size_read(inode), true);
if (err)
return err;
inode->i_mtime = inode->i_ctime = current_time(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return 0;
}
statx: Add a system call to make enhanced file info available Add a system call to make extended file information available, including file creation and some attribute flags where available through the underlying filesystem. The getattr inode operation is altered to take two additional arguments: a u32 request_mask and an unsigned int flags that indicate the synchronisation mode. This change is propagated to the vfs_getattr*() function. Functions like vfs_stat() are now inline wrappers around new functions vfs_statx() and vfs_statx_fd() to reduce stack usage. ======== OVERVIEW ======== The idea was initially proposed as a set of xattrs that could be retrieved with getxattr(), but the general preference proved to be for a new syscall with an extended stat structure. A number of requests were gathered for features to be included. The following have been included: (1) Make the fields a consistent size on all arches and make them large. (2) Spare space, request flags and information flags are provided for future expansion. (3) Better support for the y2038 problem [Arnd Bergmann] (tv_sec is an __s64). (4) Creation time: The SMB protocol carries the creation time, which could be exported by Samba, which will in turn help CIFS make use of FS-Cache as that can be used for coherency data (stx_btime). This is also specified in NFSv4 as a recommended attribute and could be exported by NFSD [Steve French]. (5) Lightweight stat: Ask for just those details of interest, and allow a netfs (such as NFS) to approximate anything not of interest, possibly without going to the server [Trond Myklebust, Ulrich Drepper, Andreas Dilger] (AT_STATX_DONT_SYNC). (6) Heavyweight stat: Force a netfs to go to the server, even if it thinks its cached attributes are up to date [Trond Myklebust] (AT_STATX_FORCE_SYNC). And the following have been left out for future extension: (7) Data version number: Could be used by userspace NFS servers [Aneesh Kumar]. Can also be used to modify fill_post_wcc() in NFSD which retrieves i_version directly, but has just called vfs_getattr(). It could get it from the kstat struct if it used vfs_xgetattr() instead. (There's disagreement on the exact semantics of a single field, since not all filesystems do this the same way). (8) BSD stat compatibility: Including more fields from the BSD stat such as creation time (st_btime) and inode generation number (st_gen) [Jeremy Allison, Bernd Schubert]. (9) Inode generation number: Useful for FUSE and userspace NFS servers [Bernd Schubert]. (This was asked for but later deemed unnecessary with the open-by-handle capability available and caused disagreement as to whether it's a security hole or not). (10) Extra coherency data may be useful in making backups [Andreas Dilger]. (No particular data were offered, but things like last backup timestamp, the data version number and the DOS archive bit would come into this category). (11) Allow the filesystem to indicate what it can/cannot provide: A filesystem can now say it doesn't support a standard stat feature if that isn't available, so if, for instance, inode numbers or UIDs don't exist or are fabricated locally... (This requires a separate system call - I have an fsinfo() call idea for this). (12) Store a 16-byte volume ID in the superblock that can be returned in struct xstat [Steve French]. (Deferred to fsinfo). (13) Include granularity fields in the time data to indicate the granularity of each of the times (NFSv4 time_delta) [Steve French]. (Deferred to fsinfo). (14) FS_IOC_GETFLAGS value. These could be translated to BSD's st_flags. Note that the Linux IOC flags are a mess and filesystems such as Ext4 define flags that aren't in linux/fs.h, so translation in the kernel may be a necessity (or, possibly, we provide the filesystem type too). (Some attributes are made available in stx_attributes, but the general feeling was that the IOC flags were to ext[234]-specific and shouldn't be exposed through statx this way). (15) Mask of features available on file (eg: ACLs, seclabel) [Brad Boyer, Michael Kerrisk]. (Deferred, probably to fsinfo. Finding out if there's an ACL or seclabal might require extra filesystem operations). (16) Femtosecond-resolution timestamps [Dave Chinner]. (A __reserved field has been left in the statx_timestamp struct for this - if there proves to be a need). (17) A set multiple attributes syscall to go with this. =============== NEW SYSTEM CALL =============== The new system call is: int ret = statx(int dfd, const char *filename, unsigned int flags, unsigned int mask, struct statx *buffer); The dfd, filename and flags parameters indicate the file to query, in a similar way to fstatat(). There is no equivalent of lstat() as that can be emulated with statx() by passing AT_SYMLINK_NOFOLLOW in flags. There is also no equivalent of fstat() as that can be emulated by passing a NULL filename to statx() with the fd of interest in dfd. Whether or not statx() synchronises the attributes with the backing store can be controlled by OR'ing a value into the flags argument (this typically only affects network filesystems): (1) AT_STATX_SYNC_AS_STAT tells statx() to behave as stat() does in this respect. (2) AT_STATX_FORCE_SYNC will require a network filesystem to synchronise its attributes with the server - which might require data writeback to occur to get the timestamps correct. (3) AT_STATX_DONT_SYNC will suppress synchronisation with the server in a network filesystem. The resulting values should be considered approximate. mask is a bitmask indicating the fields in struct statx that are of interest to the caller. The user should set this to STATX_BASIC_STATS to get the basic set returned by stat(). It should be noted that asking for more information may entail extra I/O operations. buffer points to the destination for the data. This must be 256 bytes in size. ====================== MAIN ATTRIBUTES RECORD ====================== The following structures are defined in which to return the main attribute set: struct statx_timestamp { __s64 tv_sec; __s32 tv_nsec; __s32 __reserved; }; struct statx { __u32 stx_mask; __u32 stx_blksize; __u64 stx_attributes; __u32 stx_nlink; __u32 stx_uid; __u32 stx_gid; __u16 stx_mode; __u16 __spare0[1]; __u64 stx_ino; __u64 stx_size; __u64 stx_blocks; __u64 __spare1[1]; struct statx_timestamp stx_atime; struct statx_timestamp stx_btime; struct statx_timestamp stx_ctime; struct statx_timestamp stx_mtime; __u32 stx_rdev_major; __u32 stx_rdev_minor; __u32 stx_dev_major; __u32 stx_dev_minor; __u64 __spare2[14]; }; The defined bits in request_mask and stx_mask are: STATX_TYPE Want/got stx_mode & S_IFMT STATX_MODE Want/got stx_mode & ~S_IFMT STATX_NLINK Want/got stx_nlink STATX_UID Want/got stx_uid STATX_GID Want/got stx_gid STATX_ATIME Want/got stx_atime{,_ns} STATX_MTIME Want/got stx_mtime{,_ns} STATX_CTIME Want/got stx_ctime{,_ns} STATX_INO Want/got stx_ino STATX_SIZE Want/got stx_size STATX_BLOCKS Want/got stx_blocks STATX_BASIC_STATS [The stuff in the normal stat struct] STATX_BTIME Want/got stx_btime{,_ns} STATX_ALL [All currently available stuff] stx_btime is the file creation time, stx_mask is a bitmask indicating the data provided and __spares*[] are where as-yet undefined fields can be placed. Time fields are structures with separate seconds and nanoseconds fields plus a reserved field in case we want to add even finer resolution. Note that times will be negative if before 1970; in such a case, the nanosecond fields will also be negative if not zero. The bits defined in the stx_attributes field convey information about a file, how it is accessed, where it is and what it does. The following attributes map to FS_*_FL flags and are the same numerical value: STATX_ATTR_COMPRESSED File is compressed by the fs STATX_ATTR_IMMUTABLE File is marked immutable STATX_ATTR_APPEND File is append-only STATX_ATTR_NODUMP File is not to be dumped STATX_ATTR_ENCRYPTED File requires key to decrypt in fs Within the kernel, the supported flags are listed by: KSTAT_ATTR_FS_IOC_FLAGS [Are any other IOC flags of sufficient general interest to be exposed through this interface?] New flags include: STATX_ATTR_AUTOMOUNT Object is an automount trigger These are for the use of GUI tools that might want to mark files specially, depending on what they are. Fields in struct statx come in a number of classes: (0) stx_dev_*, stx_blksize. These are local system information and are always available. (1) stx_mode, stx_nlinks, stx_uid, stx_gid, stx_[amc]time, stx_ino, stx_size, stx_blocks. These will be returned whether the caller asks for them or not. The corresponding bits in stx_mask will be set to indicate whether they actually have valid values. If the caller didn't ask for them, then they may be approximated. For example, NFS won't waste any time updating them from the server, unless as a byproduct of updating something requested. If the values don't actually exist for the underlying object (such as UID or GID on a DOS file), then the bit won't be set in the stx_mask, even if the caller asked for the value. In such a case, the returned value will be a fabrication. Note that there are instances where the type might not be valid, for instance Windows reparse points. (2) stx_rdev_*. This will be set only if stx_mode indicates we're looking at a blockdev or a chardev, otherwise will be 0. (3) stx_btime. Similar to (1), except this will be set to 0 if it doesn't exist. ======= TESTING ======= The following test program can be used to test the statx system call: samples/statx/test-statx.c Just compile and run, passing it paths to the files you want to examine. The file is built automatically if CONFIG_SAMPLES is enabled. Here's some example output. Firstly, an NFS directory that crosses to another FSID. Note that the AUTOMOUNT attribute is set because transiting this directory will cause d_automount to be invoked by the VFS. [root@andromeda ~]# /tmp/test-statx -A /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:26 Inode: 1703937 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Attributes: 0000000000001000 (-------- -------- -------- -------- -------- -------- ---m---- --------) Secondly, the result of automounting on that directory. [root@andromeda ~]# /tmp/test-statx /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:27 Inode: 2 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2017-01-31 16:46:22 +00:00
int f2fs_getattr(const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int query_flags)
{
statx: Add a system call to make enhanced file info available Add a system call to make extended file information available, including file creation and some attribute flags where available through the underlying filesystem. The getattr inode operation is altered to take two additional arguments: a u32 request_mask and an unsigned int flags that indicate the synchronisation mode. This change is propagated to the vfs_getattr*() function. Functions like vfs_stat() are now inline wrappers around new functions vfs_statx() and vfs_statx_fd() to reduce stack usage. ======== OVERVIEW ======== The idea was initially proposed as a set of xattrs that could be retrieved with getxattr(), but the general preference proved to be for a new syscall with an extended stat structure. A number of requests were gathered for features to be included. The following have been included: (1) Make the fields a consistent size on all arches and make them large. (2) Spare space, request flags and information flags are provided for future expansion. (3) Better support for the y2038 problem [Arnd Bergmann] (tv_sec is an __s64). (4) Creation time: The SMB protocol carries the creation time, which could be exported by Samba, which will in turn help CIFS make use of FS-Cache as that can be used for coherency data (stx_btime). This is also specified in NFSv4 as a recommended attribute and could be exported by NFSD [Steve French]. (5) Lightweight stat: Ask for just those details of interest, and allow a netfs (such as NFS) to approximate anything not of interest, possibly without going to the server [Trond Myklebust, Ulrich Drepper, Andreas Dilger] (AT_STATX_DONT_SYNC). (6) Heavyweight stat: Force a netfs to go to the server, even if it thinks its cached attributes are up to date [Trond Myklebust] (AT_STATX_FORCE_SYNC). And the following have been left out for future extension: (7) Data version number: Could be used by userspace NFS servers [Aneesh Kumar]. Can also be used to modify fill_post_wcc() in NFSD which retrieves i_version directly, but has just called vfs_getattr(). It could get it from the kstat struct if it used vfs_xgetattr() instead. (There's disagreement on the exact semantics of a single field, since not all filesystems do this the same way). (8) BSD stat compatibility: Including more fields from the BSD stat such as creation time (st_btime) and inode generation number (st_gen) [Jeremy Allison, Bernd Schubert]. (9) Inode generation number: Useful for FUSE and userspace NFS servers [Bernd Schubert]. (This was asked for but later deemed unnecessary with the open-by-handle capability available and caused disagreement as to whether it's a security hole or not). (10) Extra coherency data may be useful in making backups [Andreas Dilger]. (No particular data were offered, but things like last backup timestamp, the data version number and the DOS archive bit would come into this category). (11) Allow the filesystem to indicate what it can/cannot provide: A filesystem can now say it doesn't support a standard stat feature if that isn't available, so if, for instance, inode numbers or UIDs don't exist or are fabricated locally... (This requires a separate system call - I have an fsinfo() call idea for this). (12) Store a 16-byte volume ID in the superblock that can be returned in struct xstat [Steve French]. (Deferred to fsinfo). (13) Include granularity fields in the time data to indicate the granularity of each of the times (NFSv4 time_delta) [Steve French]. (Deferred to fsinfo). (14) FS_IOC_GETFLAGS value. These could be translated to BSD's st_flags. Note that the Linux IOC flags are a mess and filesystems such as Ext4 define flags that aren't in linux/fs.h, so translation in the kernel may be a necessity (or, possibly, we provide the filesystem type too). (Some attributes are made available in stx_attributes, but the general feeling was that the IOC flags were to ext[234]-specific and shouldn't be exposed through statx this way). (15) Mask of features available on file (eg: ACLs, seclabel) [Brad Boyer, Michael Kerrisk]. (Deferred, probably to fsinfo. Finding out if there's an ACL or seclabal might require extra filesystem operations). (16) Femtosecond-resolution timestamps [Dave Chinner]. (A __reserved field has been left in the statx_timestamp struct for this - if there proves to be a need). (17) A set multiple attributes syscall to go with this. =============== NEW SYSTEM CALL =============== The new system call is: int ret = statx(int dfd, const char *filename, unsigned int flags, unsigned int mask, struct statx *buffer); The dfd, filename and flags parameters indicate the file to query, in a similar way to fstatat(). There is no equivalent of lstat() as that can be emulated with statx() by passing AT_SYMLINK_NOFOLLOW in flags. There is also no equivalent of fstat() as that can be emulated by passing a NULL filename to statx() with the fd of interest in dfd. Whether or not statx() synchronises the attributes with the backing store can be controlled by OR'ing a value into the flags argument (this typically only affects network filesystems): (1) AT_STATX_SYNC_AS_STAT tells statx() to behave as stat() does in this respect. (2) AT_STATX_FORCE_SYNC will require a network filesystem to synchronise its attributes with the server - which might require data writeback to occur to get the timestamps correct. (3) AT_STATX_DONT_SYNC will suppress synchronisation with the server in a network filesystem. The resulting values should be considered approximate. mask is a bitmask indicating the fields in struct statx that are of interest to the caller. The user should set this to STATX_BASIC_STATS to get the basic set returned by stat(). It should be noted that asking for more information may entail extra I/O operations. buffer points to the destination for the data. This must be 256 bytes in size. ====================== MAIN ATTRIBUTES RECORD ====================== The following structures are defined in which to return the main attribute set: struct statx_timestamp { __s64 tv_sec; __s32 tv_nsec; __s32 __reserved; }; struct statx { __u32 stx_mask; __u32 stx_blksize; __u64 stx_attributes; __u32 stx_nlink; __u32 stx_uid; __u32 stx_gid; __u16 stx_mode; __u16 __spare0[1]; __u64 stx_ino; __u64 stx_size; __u64 stx_blocks; __u64 __spare1[1]; struct statx_timestamp stx_atime; struct statx_timestamp stx_btime; struct statx_timestamp stx_ctime; struct statx_timestamp stx_mtime; __u32 stx_rdev_major; __u32 stx_rdev_minor; __u32 stx_dev_major; __u32 stx_dev_minor; __u64 __spare2[14]; }; The defined bits in request_mask and stx_mask are: STATX_TYPE Want/got stx_mode & S_IFMT STATX_MODE Want/got stx_mode & ~S_IFMT STATX_NLINK Want/got stx_nlink STATX_UID Want/got stx_uid STATX_GID Want/got stx_gid STATX_ATIME Want/got stx_atime{,_ns} STATX_MTIME Want/got stx_mtime{,_ns} STATX_CTIME Want/got stx_ctime{,_ns} STATX_INO Want/got stx_ino STATX_SIZE Want/got stx_size STATX_BLOCKS Want/got stx_blocks STATX_BASIC_STATS [The stuff in the normal stat struct] STATX_BTIME Want/got stx_btime{,_ns} STATX_ALL [All currently available stuff] stx_btime is the file creation time, stx_mask is a bitmask indicating the data provided and __spares*[] are where as-yet undefined fields can be placed. Time fields are structures with separate seconds and nanoseconds fields plus a reserved field in case we want to add even finer resolution. Note that times will be negative if before 1970; in such a case, the nanosecond fields will also be negative if not zero. The bits defined in the stx_attributes field convey information about a file, how it is accessed, where it is and what it does. The following attributes map to FS_*_FL flags and are the same numerical value: STATX_ATTR_COMPRESSED File is compressed by the fs STATX_ATTR_IMMUTABLE File is marked immutable STATX_ATTR_APPEND File is append-only STATX_ATTR_NODUMP File is not to be dumped STATX_ATTR_ENCRYPTED File requires key to decrypt in fs Within the kernel, the supported flags are listed by: KSTAT_ATTR_FS_IOC_FLAGS [Are any other IOC flags of sufficient general interest to be exposed through this interface?] New flags include: STATX_ATTR_AUTOMOUNT Object is an automount trigger These are for the use of GUI tools that might want to mark files specially, depending on what they are. Fields in struct statx come in a number of classes: (0) stx_dev_*, stx_blksize. These are local system information and are always available. (1) stx_mode, stx_nlinks, stx_uid, stx_gid, stx_[amc]time, stx_ino, stx_size, stx_blocks. These will be returned whether the caller asks for them or not. The corresponding bits in stx_mask will be set to indicate whether they actually have valid values. If the caller didn't ask for them, then they may be approximated. For example, NFS won't waste any time updating them from the server, unless as a byproduct of updating something requested. If the values don't actually exist for the underlying object (such as UID or GID on a DOS file), then the bit won't be set in the stx_mask, even if the caller asked for the value. In such a case, the returned value will be a fabrication. Note that there are instances where the type might not be valid, for instance Windows reparse points. (2) stx_rdev_*. This will be set only if stx_mode indicates we're looking at a blockdev or a chardev, otherwise will be 0. (3) stx_btime. Similar to (1), except this will be set to 0 if it doesn't exist. ======= TESTING ======= The following test program can be used to test the statx system call: samples/statx/test-statx.c Just compile and run, passing it paths to the files you want to examine. The file is built automatically if CONFIG_SAMPLES is enabled. Here's some example output. Firstly, an NFS directory that crosses to another FSID. Note that the AUTOMOUNT attribute is set because transiting this directory will cause d_automount to be invoked by the VFS. [root@andromeda ~]# /tmp/test-statx -A /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:26 Inode: 1703937 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Attributes: 0000000000001000 (-------- -------- -------- -------- -------- -------- ---m---- --------) Secondly, the result of automounting on that directory. [root@andromeda ~]# /tmp/test-statx /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:27 Inode: 2 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2017-01-31 16:46:22 +00:00
struct inode *inode = d_inode(path->dentry);
struct f2fs_inode_info *fi = F2FS_I(inode);
unsigned int flags;
flags = fi->i_flags & (FS_FL_USER_VISIBLE | FS_PROJINHERIT_FL);
if (flags & FS_APPEND_FL)
stat->attributes |= STATX_ATTR_APPEND;
if (flags & FS_COMPR_FL)
stat->attributes |= STATX_ATTR_COMPRESSED;
if (f2fs_encrypted_inode(inode))
stat->attributes |= STATX_ATTR_ENCRYPTED;
if (flags & FS_IMMUTABLE_FL)
stat->attributes |= STATX_ATTR_IMMUTABLE;
if (flags & FS_NODUMP_FL)
stat->attributes |= STATX_ATTR_NODUMP;
stat->attributes_mask |= (STATX_ATTR_APPEND |
STATX_ATTR_COMPRESSED |
STATX_ATTR_ENCRYPTED |
STATX_ATTR_IMMUTABLE |
STATX_ATTR_NODUMP);
generic_fillattr(inode, stat);
/* we need to show initial sectors used for inline_data/dentries */
if ((S_ISREG(inode->i_mode) && f2fs_has_inline_data(inode)) ||
f2fs_has_inline_dentry(inode))
stat->blocks += (stat->size + 511) >> 9;
return 0;
}
#ifdef CONFIG_F2FS_FS_POSIX_ACL
static void __setattr_copy(struct inode *inode, const struct iattr *attr)
{
unsigned int ia_valid = attr->ia_valid;
if (ia_valid & ATTR_UID)
inode->i_uid = attr->ia_uid;
if (ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
if (ia_valid & ATTR_ATIME)
inode->i_atime = timespec_trunc(attr->ia_atime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_MTIME)
inode->i_mtime = timespec_trunc(attr->ia_mtime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_CTIME)
inode->i_ctime = timespec_trunc(attr->ia_ctime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
mode &= ~S_ISGID;
set_acl_inode(inode, mode);
}
}
#else
#define __setattr_copy setattr_copy
#endif
int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
{
struct inode *inode = d_inode(dentry);
int err;
bool size_changed = false;
if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
return -EIO;
err = setattr_prepare(dentry, attr);
if (err)
return err;
if (is_quota_modification(inode, attr)) {
err = dquot_initialize(inode);
if (err)
return err;
}
if ((attr->ia_valid & ATTR_UID &&
!uid_eq(attr->ia_uid, inode->i_uid)) ||
(attr->ia_valid & ATTR_GID &&
!gid_eq(attr->ia_gid, inode->i_gid))) {
err = dquot_transfer(inode, attr);
if (err)
return err;
}
if (attr->ia_valid & ATTR_SIZE) {
if (f2fs_encrypted_inode(inode)) {
err = fscrypt_get_encryption_info(inode);
if (err)
return err;
if (!fscrypt_has_encryption_key(inode))
return -ENOKEY;
}
if (attr->ia_size <= i_size_read(inode)) {
down_write(&F2FS_I(inode)->i_mmap_sem);
truncate_setsize(inode, attr->ia_size);
err = f2fs_truncate(inode);
up_write(&F2FS_I(inode)->i_mmap_sem);
if (err)
return err;
} else {
/*
* do not trim all blocks after i_size if target size is
* larger than i_size.
*/
down_write(&F2FS_I(inode)->i_mmap_sem);
truncate_setsize(inode, attr->ia_size);
up_write(&F2FS_I(inode)->i_mmap_sem);
/* should convert inline inode here */
if (!f2fs_may_inline_data(inode)) {
err = f2fs_convert_inline_inode(inode);
if (err)
return err;
}
inode->i_mtime = inode->i_ctime = current_time(inode);
}
down_write(&F2FS_I(inode)->i_sem);
F2FS_I(inode)->last_disk_size = i_size_read(inode);
up_write(&F2FS_I(inode)->i_sem);
size_changed = true;
}
__setattr_copy(inode, attr);
if (attr->ia_valid & ATTR_MODE) {
err = posix_acl_chmod(inode, get_inode_mode(inode));
if (err || is_inode_flag_set(inode, FI_ACL_MODE)) {
inode->i_mode = F2FS_I(inode)->i_acl_mode;
clear_inode_flag(inode, FI_ACL_MODE);
}
}
/* file size may changed here */
f2fs_mark_inode_dirty_sync(inode, size_changed);
/* inode change will produce dirty node pages flushed by checkpoint */
f2fs_balance_fs(F2FS_I_SB(inode), true);
return err;
}
const struct inode_operations f2fs_file_inode_operations = {
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.get_acl = f2fs_get_acl,
.set_acl = f2fs_set_acl,
#ifdef CONFIG_F2FS_FS_XATTR
.listxattr = f2fs_listxattr,
#endif
.fiemap = f2fs_fiemap,
};
static int fill_zero(struct inode *inode, pgoff_t index,
loff_t start, loff_t len)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct page *page;
if (!len)
return 0;
f2fs_balance_fs(sbi, true);
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
f2fs_lock_op(sbi);
page = get_new_data_page(inode, NULL, index, false);
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
f2fs_unlock_op(sbi);
if (IS_ERR(page))
return PTR_ERR(page);
f2fs_wait_on_page_writeback(page, DATA, true);
zero_user(page, start, len);
set_page_dirty(page);
f2fs_put_page(page, 1);
return 0;
}
int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
{
int err;
while (pg_start < pg_end) {
struct dnode_of_data dn;
pgoff_t end_offset, count;
set_new_dnode(&dn, inode, NULL, NULL, 0);
err = get_dnode_of_data(&dn, pg_start, LOOKUP_NODE);
if (err) {
if (err == -ENOENT) {
pg_start = get_next_page_offset(&dn, pg_start);
continue;
}
return err;
}
end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
count = min(end_offset - dn.ofs_in_node, pg_end - pg_start);
f2fs_bug_on(F2FS_I_SB(inode), count == 0 || count > end_offset);
truncate_data_blocks_range(&dn, count);
f2fs_put_dnode(&dn);
pg_start += count;
}
return 0;
}
static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
{
pgoff_t pg_start, pg_end;
loff_t off_start, off_end;
int ret;
ret = f2fs_convert_inline_inode(inode);
if (ret)
return ret;
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
pg_start = ((unsigned long long) offset) >> PAGE_SHIFT;
pg_end = ((unsigned long long) offset + len) >> PAGE_SHIFT;
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
off_start = offset & (PAGE_SIZE - 1);
off_end = (offset + len) & (PAGE_SIZE - 1);
if (pg_start == pg_end) {
ret = fill_zero(inode, pg_start, off_start,
off_end - off_start);
if (ret)
return ret;
} else {
if (off_start) {
ret = fill_zero(inode, pg_start++, off_start,
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_SIZE - off_start);
if (ret)
return ret;
}
if (off_end) {
ret = fill_zero(inode, pg_end, 0, off_end);
if (ret)
return ret;
}
if (pg_start < pg_end) {
struct address_space *mapping = inode->i_mapping;
loff_t blk_start, blk_end;
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
f2fs_balance_fs(sbi, true);
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
blk_start = (loff_t)pg_start << PAGE_SHIFT;
blk_end = (loff_t)pg_end << PAGE_SHIFT;
down_write(&F2FS_I(inode)->i_mmap_sem);
truncate_inode_pages_range(mapping, blk_start,
blk_end - 1);
f2fs: introduce a new global lock scheme In the previous version, f2fs uses global locks according to the usage types, such as directory operations, block allocation, block write, and so on. Reference the following lock types in f2fs.h. enum lock_type { RENAME, /* for renaming operations */ DENTRY_OPS, /* for directory operations */ DATA_WRITE, /* for data write */ DATA_NEW, /* for data allocation */ DATA_TRUNC, /* for data truncate */ NODE_NEW, /* for node allocation */ NODE_TRUNC, /* for node truncate */ NODE_WRITE, /* for node write */ NR_LOCK_TYPE, }; In that case, we lose the performance under the multi-threading environment, since every types of operations must be conducted one at a time. In order to address the problem, let's share the locks globally with a mutex array regardless of any types. So, let users grab a mutex and perform their jobs in parallel as much as possbile. For this, I propose a new global lock scheme as follows. 0. Data structure - f2fs_sb_info -> mutex_lock[NR_GLOBAL_LOCKS] - f2fs_sb_info -> node_write 1. mutex_lock_op(sbi) - try to get an avaiable lock from the array. - returns the index of the gottern lock variable. 2. mutex_unlock_op(sbi, index of the lock) - unlock the given index of the lock. 3. mutex_lock_all(sbi) - grab all the locks in the array before the checkpoint. 4. mutex_unlock_all(sbi) - release all the locks in the array after checkpoint. 5. block_operations() - call mutex_lock_all() - sync_dirty_dir_inodes() - grab node_write - sync_node_pages() Note that, the pairs of mutex_lock_op()/mutex_unlock_op() and mutex_lock_all()/mutex_unlock_all() should be used together. Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-11-22 07:21:29 +00:00
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
f2fs_lock_op(sbi);
ret = truncate_hole(inode, pg_start, pg_end);
f2fs: use rw_sem instead of fs_lock(locks mutex) The fs_locks is used to block other ops(ex, recovery) when doing checkpoint. And each other operate routine(besides checkpoint) needs to acquire a fs_lock, there is a terrible problem here, if these are too many concurrency threads acquiring fs_lock, so that they will block each other and may lead to some performance problem, but this is not the phenomenon we want to see. Though there are some optimization patches introduced to enhance the usage of fs_lock, but the thorough solution is using a *rw_sem* to replace the fs_lock. Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other, this can avoid the problem described above completely. Because of the weakness of rw_sem, the above change may introduce a potential problem that the checkpoint thread might get starved if other threads are intensively locking the read semaphore for I/O.(Pointed out by Xu Jin) In order to avoid this, a wait_list is introduced, the appending read semaphore ops will be dropped into the wait_list if checkpoint thread is waiting for write semaphore, and will be waked up when checkpoint thread gives up write semaphore. Thanks to Kim's previous review and test, and will be very glad to see other guys' performance tests about this patch. V2: -fix the potential starvation problem. -use more suitable func name suggested by Xu Jin. Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> [Jaegeuk Kim: adjust minor coding standard] Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-09-27 10:08:30 +00:00
f2fs_unlock_op(sbi);
up_write(&F2FS_I(inode)->i_mmap_sem);
}
}
return ret;
}
static int __read_out_blkaddrs(struct inode *inode, block_t *blkaddr,
int *do_replace, pgoff_t off, pgoff_t len)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct dnode_of_data dn;
int ret, done, i;
next_dnode:
set_new_dnode(&dn, inode, NULL, NULL, 0);
ret = get_dnode_of_data(&dn, off, LOOKUP_NODE_RA);
if (ret && ret != -ENOENT) {
return ret;
} else if (ret == -ENOENT) {
if (dn.max_level == 0)
return -ENOENT;
done = min((pgoff_t)ADDRS_PER_BLOCK - dn.ofs_in_node, len);
blkaddr += done;
do_replace += done;
goto next;
}
done = min((pgoff_t)ADDRS_PER_PAGE(dn.node_page, inode) -
dn.ofs_in_node, len);
for (i = 0; i < done; i++, blkaddr++, do_replace++, dn.ofs_in_node++) {
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
*blkaddr = datablock_addr(dn.inode,
dn.node_page, dn.ofs_in_node);
if (!is_checkpointed_data(sbi, *blkaddr)) {
if (test_opt(sbi, LFS)) {
f2fs_put_dnode(&dn);
return -ENOTSUPP;
}
/* do not invalidate this block address */
f2fs_update_data_blkaddr(&dn, NULL_ADDR);
*do_replace = 1;
}
}
f2fs_put_dnode(&dn);
next:
len -= done;
off += done;
if (len)
goto next_dnode;
return 0;
}
static int __roll_back_blkaddrs(struct inode *inode, block_t *blkaddr,
int *do_replace, pgoff_t off, int len)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct dnode_of_data dn;
int ret, i;
for (i = 0; i < len; i++, do_replace++, blkaddr++) {
if (*do_replace == 0)
continue;
set_new_dnode(&dn, inode, NULL, NULL, 0);
ret = get_dnode_of_data(&dn, off + i, LOOKUP_NODE_RA);
if (ret) {
dec_valid_block_count(sbi, inode, 1);
invalidate_blocks(sbi, *blkaddr);
} else {
f2fs_update_data_blkaddr(&dn, *blkaddr);
}
f2fs_put_dnode(&dn);
}
return 0;
}
static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode,
block_t *blkaddr, int *do_replace,
pgoff_t src, pgoff_t dst, pgoff_t len, bool full)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(src_inode);
pgoff_t i = 0;
int ret;
while (i < len) {
if (blkaddr[i] == NULL_ADDR && !full) {
i++;
continue;
}
if (do_replace[i] || blkaddr[i] == NULL_ADDR) {
struct dnode_of_data dn;
struct node_info ni;
size_t new_size;
pgoff_t ilen;
set_new_dnode(&dn, dst_inode, NULL, NULL, 0);
ret = get_dnode_of_data(&dn, dst + i, ALLOC_NODE);
if (ret)
return ret;
get_node_info(sbi, dn.nid, &ni);
ilen = min((pgoff_t)
ADDRS_PER_PAGE(dn.node_page, dst_inode) -
dn.ofs_in_node, len - i);
do {
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
dn.data_blkaddr = datablock_addr(dn.inode,
dn.node_page, dn.ofs_in_node);
truncate_data_blocks_range(&dn, 1);
if (do_replace[i]) {
f2fs_i_blocks_write(src_inode,
1, false, false);
f2fs_i_blocks_write(dst_inode,
1, true, false);
f2fs_replace_block(sbi, &dn, dn.data_blkaddr,
blkaddr[i], ni.version, true, false);
do_replace[i] = 0;
}
dn.ofs_in_node++;
i++;
new_size = (dst + i) << PAGE_SHIFT;
if (dst_inode->i_size < new_size)
f2fs_i_size_write(dst_inode, new_size);
} while (--ilen && (do_replace[i] || blkaddr[i] == NULL_ADDR));
f2fs_put_dnode(&dn);
} else {
struct page *psrc, *pdst;
psrc = get_lock_data_page(src_inode, src + i, true);
if (IS_ERR(psrc))
return PTR_ERR(psrc);
pdst = get_new_data_page(dst_inode, NULL, dst + i,
true);
if (IS_ERR(pdst)) {
f2fs_put_page(psrc, 1);
return PTR_ERR(pdst);
}
f2fs_copy_page(psrc, pdst);
set_page_dirty(pdst);
f2fs_put_page(pdst, 1);
f2fs_put_page(psrc, 1);
ret = truncate_hole(src_inode, src + i, src + i + 1);
if (ret)
return ret;
i++;
}
}
return 0;
}
static int __exchange_data_block(struct inode *src_inode,
struct inode *dst_inode, pgoff_t src, pgoff_t dst,
pgoff_t len, bool full)
{
block_t *src_blkaddr;
int *do_replace;
pgoff_t olen;
int ret;
while (len) {
olen = min((pgoff_t)4 * ADDRS_PER_BLOCK, len);
mm: introduce kv[mz]alloc helpers Patch series "kvmalloc", v5. There are many open coded kmalloc with vmalloc fallback instances in the tree. Most of them are not careful enough or simply do not care about the underlying semantic of the kmalloc/page allocator which means that a) some vmalloc fallbacks are basically unreachable because the kmalloc part will keep retrying until it succeeds b) the page allocator can invoke a really disruptive steps like the OOM killer to move forward which doesn't sound appropriate when we consider that the vmalloc fallback is available. As it can be seen implementing kvmalloc requires quite an intimate knowledge if the page allocator and the memory reclaim internals which strongly suggests that a helper should be implemented in the memory subsystem proper. Most callers, I could find, have been converted to use the helper instead. This is patch 6. There are some more relying on __GFP_REPEAT in the networking stack which I have converted as well and Eric Dumazet was not opposed [2] to convert them as well. [1] http://lkml.kernel.org/r/20170130094940.13546-1-mhocko@kernel.org [2] http://lkml.kernel.org/r/1485273626.16328.301.camel@edumazet-glaptop3.roam.corp.google.com This patch (of 9): Using kmalloc with the vmalloc fallback for larger allocations is a common pattern in the kernel code. Yet we do not have any common helper for that and so users have invented their own helpers. Some of them are really creative when doing so. Let's just add kv[mz]alloc and make sure it is implemented properly. This implementation makes sure to not make a large memory pressure for > PAGE_SZE requests (__GFP_NORETRY) and also to not warn about allocation failures. This also rules out the OOM killer as the vmalloc is a more approapriate fallback than a disruptive user visible action. This patch also changes some existing users and removes helpers which are specific for them. In some cases this is not possible (e.g. ext4_kvmalloc, libcfs_kvzalloc) because those seems to be broken and require GFP_NO{FS,IO} context which is not vmalloc compatible in general (note that the page table allocation is GFP_KERNEL). Those need to be fixed separately. While we are at it, document that __vmalloc{_node} about unsupported gfp mask because there seems to be a lot of confusion out there. kvmalloc_node will warn about GFP_KERNEL incompatible (which are not superset) flags to catch new abusers. Existing ones would have to die slowly. [sfr@canb.auug.org.au: f2fs fixup] Link: http://lkml.kernel.org/r/20170320163735.332e64b7@canb.auug.org.au Link: http://lkml.kernel.org/r/20170306103032.2540-2-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Reviewed-by: Andreas Dilger <adilger@dilger.ca> [ext4 part] Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: John Hubbard <jhubbard@nvidia.com> Cc: David Miller <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-08 22:57:09 +00:00
src_blkaddr = kvzalloc(sizeof(block_t) * olen, GFP_KERNEL);
if (!src_blkaddr)
return -ENOMEM;
mm: introduce kv[mz]alloc helpers Patch series "kvmalloc", v5. There are many open coded kmalloc with vmalloc fallback instances in the tree. Most of them are not careful enough or simply do not care about the underlying semantic of the kmalloc/page allocator which means that a) some vmalloc fallbacks are basically unreachable because the kmalloc part will keep retrying until it succeeds b) the page allocator can invoke a really disruptive steps like the OOM killer to move forward which doesn't sound appropriate when we consider that the vmalloc fallback is available. As it can be seen implementing kvmalloc requires quite an intimate knowledge if the page allocator and the memory reclaim internals which strongly suggests that a helper should be implemented in the memory subsystem proper. Most callers, I could find, have been converted to use the helper instead. This is patch 6. There are some more relying on __GFP_REPEAT in the networking stack which I have converted as well and Eric Dumazet was not opposed [2] to convert them as well. [1] http://lkml.kernel.org/r/20170130094940.13546-1-mhocko@kernel.org [2] http://lkml.kernel.org/r/1485273626.16328.301.camel@edumazet-glaptop3.roam.corp.google.com This patch (of 9): Using kmalloc with the vmalloc fallback for larger allocations is a common pattern in the kernel code. Yet we do not have any common helper for that and so users have invented their own helpers. Some of them are really creative when doing so. Let's just add kv[mz]alloc and make sure it is implemented properly. This implementation makes sure to not make a large memory pressure for > PAGE_SZE requests (__GFP_NORETRY) and also to not warn about allocation failures. This also rules out the OOM killer as the vmalloc is a more approapriate fallback than a disruptive user visible action. This patch also changes some existing users and removes helpers which are specific for them. In some cases this is not possible (e.g. ext4_kvmalloc, libcfs_kvzalloc) because those seems to be broken and require GFP_NO{FS,IO} context which is not vmalloc compatible in general (note that the page table allocation is GFP_KERNEL). Those need to be fixed separately. While we are at it, document that __vmalloc{_node} about unsupported gfp mask because there seems to be a lot of confusion out there. kvmalloc_node will warn about GFP_KERNEL incompatible (which are not superset) flags to catch new abusers. Existing ones would have to die slowly. [sfr@canb.auug.org.au: f2fs fixup] Link: http://lkml.kernel.org/r/20170320163735.332e64b7@canb.auug.org.au Link: http://lkml.kernel.org/r/20170306103032.2540-2-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Reviewed-by: Andreas Dilger <adilger@dilger.ca> [ext4 part] Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: John Hubbard <jhubbard@nvidia.com> Cc: David Miller <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-05-08 22:57:09 +00:00
do_replace = kvzalloc(sizeof(int) * olen, GFP_KERNEL);
if (!do_replace) {
kvfree(src_blkaddr);
return -ENOMEM;
}
ret = __read_out_blkaddrs(src_inode, src_blkaddr,
do_replace, src, olen);
if (ret)
goto roll_back;
ret = __clone_blkaddrs(src_inode, dst_inode, src_blkaddr,
do_replace, src, dst, olen, full);
if (ret)
goto roll_back;
src += olen;
dst += olen;
len -= olen;
kvfree(src_blkaddr);
kvfree(do_replace);
}
return 0;
roll_back:
__roll_back_blkaddrs(src_inode, src_blkaddr, do_replace, src, len);
kvfree(src_blkaddr);
kvfree(do_replace);
return ret;
}
static int f2fs_do_collapse(struct inode *inode, pgoff_t start, pgoff_t end)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
pgoff_t nrpages = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
int ret;
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
f2fs_drop_extent_tree(inode);
ret = __exchange_data_block(inode, inode, end, start, nrpages - end, true);
f2fs_unlock_op(sbi);
return ret;
}
static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
{
pgoff_t pg_start, pg_end;
loff_t new_size;
int ret;
if (offset + len >= i_size_read(inode))
return -EINVAL;
/* collapse range should be aligned to block size of f2fs. */
if (offset & (F2FS_BLKSIZE - 1) || len & (F2FS_BLKSIZE - 1))
return -EINVAL;
ret = f2fs_convert_inline_inode(inode);
if (ret)
return ret;
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
pg_start = offset >> PAGE_SHIFT;
pg_end = (offset + len) >> PAGE_SHIFT;
/* avoid gc operation during block exchange */
down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
down_write(&F2FS_I(inode)->i_mmap_sem);
/* write out all dirty pages from offset */
ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
if (ret)
goto out_unlock;
truncate_pagecache(inode, offset);
ret = f2fs_do_collapse(inode, pg_start, pg_end);
if (ret)
goto out_unlock;
/* write out all moved pages, if possible */
filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
truncate_pagecache(inode, offset);
new_size = i_size_read(inode) - len;
truncate_pagecache(inode, new_size);
ret = truncate_blocks(inode, new_size, true);
if (!ret)
f2fs_i_size_write(inode, new_size);
out_unlock:
up_write(&F2FS_I(inode)->i_mmap_sem);
up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
return ret;
}
static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start,
pgoff_t end)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
pgoff_t index = start;
unsigned int ofs_in_node = dn->ofs_in_node;
blkcnt_t count = 0;
int ret;
for (; index < end; index++, dn->ofs_in_node++) {
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
if (datablock_addr(dn->inode, dn->node_page,
dn->ofs_in_node) == NULL_ADDR)
count++;
}
dn->ofs_in_node = ofs_in_node;
ret = reserve_new_blocks(dn, count);
if (ret)
return ret;
dn->ofs_in_node = ofs_in_node;
for (index = start; index < end; index++, dn->ofs_in_node++) {
f2fs: enhance on-disk inode structure scalability This patch add new flag F2FS_EXTRA_ATTR storing in inode.i_inline to indicate that on-disk structure of current inode is extended. In order to extend, we changed the inode structure a bit: Original one: struct f2fs_inode { ... struct f2fs_extent i_ext; __le32 i_addr[DEF_ADDRS_PER_INODE]; __le32 i_nid[DEF_NIDS_PER_INODE]; } Extended one: struct f2fs_inode { ... struct f2fs_extent i_ext; union { struct { __le16 i_extra_isize; __le16 i_padding; __le32 i_extra_end[0]; }; __le32 i_addr[DEF_ADDRS_PER_INODE]; }; __le32 i_nid[DEF_NIDS_PER_INODE]; } Once F2FS_EXTRA_ATTR is set, we will steal four bytes in the head of i_addr field for storing i_extra_isize and i_padding. with i_extra_isize, we can calculate actual size of reserved space in i_addr, available attribute fields included in total extra attribute fields for current inode can be described as below: +--------------------+ | .i_mode | | ... | | .i_ext | +--------------------+ | .i_extra_isize |-----+ | .i_padding | | | .i_prjid | | | .i_atime_extra | | | .i_ctime_extra | | | .i_mtime_extra |<----+ | .i_inode_cs |<----- store blkaddr/inline from here | .i_xattr_cs | | ... | +--------------------+ | | | block address | | | +--------------------+ | .i_nid | +--------------------+ | node_footer | | (nid, ino, offset) | +--------------------+ Hence, with this patch, we would enhance scalability of f2fs inode for storing more newly added attribute. Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2017-07-18 16:19:06 +00:00
dn->data_blkaddr = datablock_addr(dn->inode,
dn->node_page, dn->ofs_in_node);
/*
* reserve_new_blocks will not guarantee entire block
* allocation.
*/
if (dn->data_blkaddr == NULL_ADDR) {
ret = -ENOSPC;
break;
}
if (dn->data_blkaddr != NEW_ADDR) {
invalidate_blocks(sbi, dn->data_blkaddr);
dn->data_blkaddr = NEW_ADDR;
set_data_blkaddr(dn);
}
}
f2fs_update_extent_cache_range(dn, start, 0, index - start);
return ret;
}
static int f2fs_zero_range(struct inode *inode, loff_t offset, loff_t len,
int mode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct address_space *mapping = inode->i_mapping;
pgoff_t index, pg_start, pg_end;
loff_t new_size = i_size_read(inode);
loff_t off_start, off_end;
int ret = 0;
ret = inode_newsize_ok(inode, (len + offset));
if (ret)
return ret;
ret = f2fs_convert_inline_inode(inode);
if (ret)
return ret;
down_write(&F2FS_I(inode)->i_mmap_sem);
ret = filemap_write_and_wait_range(mapping, offset, offset + len - 1);
if (ret)
goto out_sem;
truncate_pagecache_range(inode, offset, offset + len - 1);
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
pg_start = ((unsigned long long) offset) >> PAGE_SHIFT;
pg_end = ((unsigned long long) offset + len) >> PAGE_SHIFT;
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
off_start = offset & (PAGE_SIZE - 1);
off_end = (offset + len) & (PAGE_SIZE - 1);
if (pg_start == pg_end) {
ret = fill_zero(inode, pg_start, off_start,
off_end - off_start);
if (ret)
goto out_sem;
new_size = max_t(loff_t, new_size, offset + len);
} else {
if (off_start) {
ret = fill_zero(inode, pg_start++, off_start,
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_SIZE - off_start);
if (ret)
goto out_sem;
new_size = max_t(loff_t, new_size,
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
(loff_t)pg_start << PAGE_SHIFT);
}
for (index = pg_start; index < pg_end;) {
struct dnode_of_data dn;
unsigned int end_offset;
pgoff_t end;
f2fs_lock_op(sbi);
set_new_dnode(&dn, inode, NULL, NULL, 0);
ret = get_dnode_of_data(&dn, index, ALLOC_NODE);
if (ret) {
f2fs_unlock_op(sbi);
goto out;
}
end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
end = min(pg_end, end_offset - dn.ofs_in_node + index);
ret = f2fs_do_zero_range(&dn, index, end);
f2fs_put_dnode(&dn);
f2fs_unlock_op(sbi);
f2fs_balance_fs(sbi, dn.node_changed);
if (ret)
goto out;
index = end;
new_size = max_t(loff_t, new_size,
(loff_t)index << PAGE_SHIFT);
}
if (off_end) {
ret = fill_zero(inode, pg_end, 0, off_end);
if (ret)
goto out;
new_size = max_t(loff_t, new_size, offset + len);
}
}
out:
if (!(mode & FALLOC_FL_KEEP_SIZE) && i_size_read(inode) < new_size)
f2fs_i_size_write(inode, new_size);
out_sem:
up_write(&F2FS_I(inode)->i_mmap_sem);
return ret;
}
static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
pgoff_t nr, pg_start, pg_end, delta, idx;
loff_t new_size;
int ret = 0;
new_size = i_size_read(inode) + len;
ret = inode_newsize_ok(inode, new_size);
if (ret)
return ret;
if (offset >= i_size_read(inode))
return -EINVAL;
/* insert range should be aligned to block size of f2fs. */
if (offset & (F2FS_BLKSIZE - 1) || len & (F2FS_BLKSIZE - 1))
return -EINVAL;
ret = f2fs_convert_inline_inode(inode);
if (ret)
return ret;
f2fs_balance_fs(sbi, true);
/* avoid gc operation during block exchange */
down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
down_write(&F2FS_I(inode)->i_mmap_sem);
ret = truncate_blocks(inode, i_size_read(inode), true);
if (ret)
goto out;
/* write out all dirty pages from offset */
ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
if (ret)
goto out;
truncate_pagecache(inode, offset);
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
pg_start = offset >> PAGE_SHIFT;
pg_end = (offset + len) >> PAGE_SHIFT;
delta = pg_end - pg_start;
idx = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
while (!ret && idx > pg_start) {
nr = idx - pg_start;
if (nr > delta)
nr = delta;
idx -= nr;
f2fs_lock_op(sbi);
f2fs_drop_extent_tree(inode);
ret = __exchange_data_block(inode, inode, idx,
idx + delta, nr, false);
f2fs_unlock_op(sbi);
}
/* write out all moved pages, if possible */
filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
truncate_pagecache(inode, offset);
if (!ret)
f2fs_i_size_write(inode, new_size);
out:
up_write(&F2FS_I(inode)->i_mmap_sem);
up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
return ret;
}
static int expand_inode_data(struct inode *inode, loff_t offset,
loff_t len, int mode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
pgoff_t pg_end;
loff_t new_size = i_size_read(inode);
loff_t off_end;
int err;
err = inode_newsize_ok(inode, (len + offset));
if (err)
return err;
err = f2fs_convert_inline_inode(inode);
if (err)
return err;
f2fs_balance_fs(sbi, true);
pg_end = ((unsigned long long)offset + len) >> PAGE_SHIFT;
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
off_end = (offset + len) & (PAGE_SIZE - 1);
map.m_lblk = ((unsigned long long)offset) >> PAGE_SHIFT;
map.m_len = pg_end - map.m_lblk;
if (off_end)
map.m_len++;
err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO);
if (err) {
pgoff_t last_off;
if (!map.m_len)
return err;
last_off = map.m_lblk + map.m_len - 1;
/* update new size to the failed position */
new_size = (last_off == pg_end) ? offset + len:
(loff_t)(last_off + 1) << PAGE_SHIFT;
} else {
new_size = ((loff_t)pg_end << PAGE_SHIFT) + off_end;
}
if (new_size > i_size_read(inode)) {
if (mode & FALLOC_FL_KEEP_SIZE)
file_set_keep_isize(inode);
else
f2fs_i_size_write(inode, new_size);
}
return err;
}
static long f2fs_fallocate(struct file *file, int mode,
loff_t offset, loff_t len)
{
struct inode *inode = file_inode(file);
long ret = 0;
if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
return -EIO;
/* f2fs only support ->fallocate for regular file */
if (!S_ISREG(inode->i_mode))
return -EINVAL;
if (f2fs_encrypted_inode(inode) &&
(mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE)))
return -EOPNOTSUPP;
if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE |
FALLOC_FL_INSERT_RANGE))
return -EOPNOTSUPP;
inode_lock(inode);
if (mode & FALLOC_FL_PUNCH_HOLE) {
if (offset >= inode->i_size)
goto out;
ret = punch_hole(inode, offset, len);
} else if (mode & FALLOC_FL_COLLAPSE_RANGE) {
ret = f2fs_collapse_range(inode, offset, len);
} else if (mode & FALLOC_FL_ZERO_RANGE) {
ret = f2fs_zero_range(inode, offset, len, mode);
} else if (mode & FALLOC_FL_INSERT_RANGE) {
ret = f2fs_insert_range(inode, offset, len);
} else {
ret = expand_inode_data(inode, offset, len, mode);
}
if (!ret) {
inode->i_mtime = inode->i_ctime = current_time(inode);
f2fs_mark_inode_dirty_sync(inode, false);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
}
out:
inode_unlock(inode);
trace_f2fs_fallocate(inode, mode, offset, len, ret);
return ret;
}
static int f2fs_release_file(struct inode *inode, struct file *filp)
{
/*
* f2fs_relase_file is called at every close calls. So we should
* not drop any inmemory pages by close called by other process.
*/
if (!(filp->f_mode & FMODE_WRITE) ||
atomic_read(&inode->i_writecount) != 1)
return 0;
/* some remained atomic pages should discarded */
if (f2fs_is_atomic_file(inode))
drop_inmem_pages(inode);
if (f2fs_is_volatile_file(inode)) {
clear_inode_flag(inode, FI_VOLATILE_FILE);
stat_dec_volatile_write(inode);
set_inode_flag(inode, FI_DROP_CACHE);
filemap_fdatawrite(inode->i_mapping);
clear_inode_flag(inode, FI_DROP_CACHE);
}
return 0;
}
static int f2fs_file_flush(struct file *file, fl_owner_t id)
{
struct inode *inode = file_inode(file);
/*
* If the process doing a transaction is crashed, we should do
* roll-back. Otherwise, other reader/write can see corrupted database
* until all the writers close its file. Since this should be done
* before dropping file lock, it needs to do in ->flush.
*/
if (f2fs_is_atomic_file(inode) &&
F2FS_I(inode)->inmem_task == current)
drop_inmem_pages(inode);
return 0;
}
static int f2fs_ioc_getflags(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_inode_info *fi = F2FS_I(inode);
unsigned int flags = fi->i_flags &
(FS_FL_USER_VISIBLE | FS_PROJINHERIT_FL);
return put_user(flags, (int __user *)arg);
}
static int __f2fs_ioc_setflags(struct inode *inode, unsigned int flags)
{
struct f2fs_inode_info *fi = F2FS_I(inode);
unsigned int oldflags;
/* Is it quota file? Do not allow user to mess with it */
if (IS_NOQUOTA(inode))
return -EPERM;
flags = f2fs_mask_flags(inode->i_mode, flags);
oldflags = fi->i_flags;
if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL))
if (!capable(CAP_LINUX_IMMUTABLE))
return -EPERM;
flags = flags & (FS_FL_USER_MODIFIABLE | FS_PROJINHERIT_FL);
flags |= oldflags & ~(FS_FL_USER_MODIFIABLE | FS_PROJINHERIT_FL);
fi->i_flags = flags;
if (fi->i_flags & FS_PROJINHERIT_FL)
set_inode_flag(inode, FI_PROJ_INHERIT);
else
clear_inode_flag(inode, FI_PROJ_INHERIT);
inode->i_ctime = current_time(inode);
f2fs_set_inode_flags(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return 0;
}
static int f2fs_ioc_setflags(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
unsigned int flags;
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
if (get_user(flags, (int __user *)arg))
return -EFAULT;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
ret = __f2fs_ioc_setflags(inode, flags);
inode_unlock(inode);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_getversion(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
return put_user(inode->i_generation, (int __user *)arg);
}
static int f2fs_ioc_start_atomic_write(struct file *filp)
{
struct inode *inode = file_inode(filp);
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
if (!S_ISREG(inode->i_mode))
return -EINVAL;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
if (f2fs_is_atomic_file(inode))
goto out;
ret = f2fs_convert_inline_inode(inode);
if (ret)
goto out;
set_inode_flag(inode, FI_ATOMIC_FILE);
set_inode_flag(inode, FI_HOT_DATA);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
if (!get_dirty_pages(inode))
goto inc_stat;
f2fs_msg(F2FS_I_SB(inode)->sb, KERN_WARNING,
"Unexpected flush for atomic writes: ino=%lu, npages=%u",
inode->i_ino, get_dirty_pages(inode));
ret = filemap_write_and_wait_range(inode->i_mapping, 0, LLONG_MAX);
if (ret) {
clear_inode_flag(inode, FI_ATOMIC_FILE);
clear_inode_flag(inode, FI_HOT_DATA);
goto out;
}
inc_stat:
F2FS_I(inode)->inmem_task = current;
stat_inc_atomic_write(inode);
stat_update_max_atomic_write(inode);
out:
inode_unlock(inode);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_commit_atomic_write(struct file *filp)
{
struct inode *inode = file_inode(filp);
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
if (f2fs_is_volatile_file(inode))
goto err_out;
f2fs: call set_page_dirty to attach i_wb for cgroup The cgroup attaches inode->i_wb via mark_inode_dirty and when set_page_writeback is called, __inc_wb_stat() updates i_wb's stat. So, we need to explicitly call set_page_dirty->__mark_inode_dirty in prior to any writebacking pages. This patch should resolve the following kernel panic reported by Andreas Reis. https://bugzilla.kernel.org/show_bug.cgi?id=101801 --- Comment #2 from Andreas Reis <andreas.reis@gmail.com> --- BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8 IP: [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 PGD 2951ff067 PUD 2df43f067 PMD 0 Oops: 0000 [#1] PREEMPT SMP Modules linked in: CPU: 7 PID: 10356 Comm: gcc Tainted: G W 4.2.0-1-cu #1 Hardware name: Gigabyte Technology Co., Ltd. G1.Sniper M5/G1.Sniper M5, BIOS T01 02/03/2015 task: ffff880295044f80 ti: ffff880295140000 task.ti: ffff880295140000 RIP: 0010:[<ffffffff8149deea>] [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 RSP: 0018:ffff880295143ac8 EFLAGS: 00010082 RAX: 0000000000000003 RBX: ffffea000a526d40 RCX: 0000000000000001 RDX: 0000000000000020 RSI: 0000000000000001 RDI: 0000000000000088 RBP: ffff880295143ae8 R08: 0000000000000000 R09: ffff88008f69bb30 R10: 00000000fffffffa R11: 0000000000000000 R12: 0000000000000088 R13: 0000000000000001 R14: ffff88041d099000 R15: ffff880084a205d0 FS: 00007f8549374700(0000) GS:ffff88042f3c0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000000000a8 CR3: 000000033e1d5000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Stack: 0000000000000000 ffffea000a526d40 ffff880084a20738 ffff880084a20750 ffff880295143b48 ffffffff811cc91e ffff880000000000 0000000000000296 0000000000000000 ffff880417090198 0000000000000000 ffffea000a526d40 Call Trace: [<ffffffff811cc91e>] __test_set_page_writeback+0xde/0x1d0 [<ffffffff813fee87>] do_write_data_page+0xe7/0x3a0 [<ffffffff813faeea>] gc_data_segment+0x5aa/0x640 [<ffffffff813fb0b8>] do_garbage_collect+0x138/0x150 [<ffffffff813fb3fe>] f2fs_gc+0x1be/0x3e0 [<ffffffff81405541>] f2fs_balance_fs+0x81/0x90 [<ffffffff813ee357>] f2fs_unlink+0x47/0x1d0 [<ffffffff81239329>] vfs_unlink+0x109/0x1b0 [<ffffffff8123e3d7>] do_unlinkat+0x287/0x2c0 [<ffffffff8123ebc6>] SyS_unlink+0x16/0x20 [<ffffffff81942e2e>] entry_SYSCALL_64_fastpath+0x12/0x71 Code: 41 5e 5d c3 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 55 49 89 f5 41 54 49 89 fc 53 48 83 ec 08 65 ff 05 e6 d9 b6 7e <48> 8b 47 20 48 63 ca 65 8b 18 48 63 db 48 01 f3 48 39 cb 7d 0a RIP [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 RSP <ffff880295143ac8> CR2: 00000000000000a8 ---[ end trace 5132449a58ed93a3 ]--- note: gcc[10356] exited with preempt_count 2 Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-07-25 07:29:17 +00:00
if (f2fs_is_atomic_file(inode)) {
ret = commit_inmem_pages(inode);
if (ret)
goto err_out;
ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true);
if (!ret) {
clear_inode_flag(inode, FI_ATOMIC_FILE);
clear_inode_flag(inode, FI_HOT_DATA);
stat_dec_atomic_write(inode);
}
} else {
ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 1, false);
f2fs: call set_page_dirty to attach i_wb for cgroup The cgroup attaches inode->i_wb via mark_inode_dirty and when set_page_writeback is called, __inc_wb_stat() updates i_wb's stat. So, we need to explicitly call set_page_dirty->__mark_inode_dirty in prior to any writebacking pages. This patch should resolve the following kernel panic reported by Andreas Reis. https://bugzilla.kernel.org/show_bug.cgi?id=101801 --- Comment #2 from Andreas Reis <andreas.reis@gmail.com> --- BUG: unable to handle kernel NULL pointer dereference at 00000000000000a8 IP: [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 PGD 2951ff067 PUD 2df43f067 PMD 0 Oops: 0000 [#1] PREEMPT SMP Modules linked in: CPU: 7 PID: 10356 Comm: gcc Tainted: G W 4.2.0-1-cu #1 Hardware name: Gigabyte Technology Co., Ltd. G1.Sniper M5/G1.Sniper M5, BIOS T01 02/03/2015 task: ffff880295044f80 ti: ffff880295140000 task.ti: ffff880295140000 RIP: 0010:[<ffffffff8149deea>] [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 RSP: 0018:ffff880295143ac8 EFLAGS: 00010082 RAX: 0000000000000003 RBX: ffffea000a526d40 RCX: 0000000000000001 RDX: 0000000000000020 RSI: 0000000000000001 RDI: 0000000000000088 RBP: ffff880295143ae8 R08: 0000000000000000 R09: ffff88008f69bb30 R10: 00000000fffffffa R11: 0000000000000000 R12: 0000000000000088 R13: 0000000000000001 R14: ffff88041d099000 R15: ffff880084a205d0 FS: 00007f8549374700(0000) GS:ffff88042f3c0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000000000a8 CR3: 000000033e1d5000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Stack: 0000000000000000 ffffea000a526d40 ffff880084a20738 ffff880084a20750 ffff880295143b48 ffffffff811cc91e ffff880000000000 0000000000000296 0000000000000000 ffff880417090198 0000000000000000 ffffea000a526d40 Call Trace: [<ffffffff811cc91e>] __test_set_page_writeback+0xde/0x1d0 [<ffffffff813fee87>] do_write_data_page+0xe7/0x3a0 [<ffffffff813faeea>] gc_data_segment+0x5aa/0x640 [<ffffffff813fb0b8>] do_garbage_collect+0x138/0x150 [<ffffffff813fb3fe>] f2fs_gc+0x1be/0x3e0 [<ffffffff81405541>] f2fs_balance_fs+0x81/0x90 [<ffffffff813ee357>] f2fs_unlink+0x47/0x1d0 [<ffffffff81239329>] vfs_unlink+0x109/0x1b0 [<ffffffff8123e3d7>] do_unlinkat+0x287/0x2c0 [<ffffffff8123ebc6>] SyS_unlink+0x16/0x20 [<ffffffff81942e2e>] entry_SYSCALL_64_fastpath+0x12/0x71 Code: 41 5e 5d c3 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 55 49 89 f5 41 54 49 89 fc 53 48 83 ec 08 65 ff 05 e6 d9 b6 7e <48> 8b 47 20 48 63 ca 65 8b 18 48 63 db 48 01 f3 48 39 cb 7d 0a RIP [<ffffffff8149deea>] __percpu_counter_add+0x1a/0x90 RSP <ffff880295143ac8> CR2: 00000000000000a8 ---[ end trace 5132449a58ed93a3 ]--- note: gcc[10356] exited with preempt_count 2 Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2015-07-25 07:29:17 +00:00
}
err_out:
inode_unlock(inode);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_start_volatile_write(struct file *filp)
{
struct inode *inode = file_inode(filp);
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
if (!S_ISREG(inode->i_mode))
return -EINVAL;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
if (f2fs_is_volatile_file(inode))
goto out;
ret = f2fs_convert_inline_inode(inode);
if (ret)
goto out;
stat_inc_volatile_write(inode);
stat_update_max_volatile_write(inode);
set_inode_flag(inode, FI_VOLATILE_FILE);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
out:
inode_unlock(inode);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_release_volatile_write(struct file *filp)
{
struct inode *inode = file_inode(filp);
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
if (!f2fs_is_volatile_file(inode))
goto out;
if (!f2fs_is_first_block_written(inode)) {
ret = truncate_partial_data_page(inode, 0, true);
goto out;
}
ret = punch_hole(inode, 0, F2FS_BLKSIZE);
out:
inode_unlock(inode);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_abort_volatile_write(struct file *filp)
{
struct inode *inode = file_inode(filp);
int ret;
if (!inode_owner_or_capable(inode))
return -EACCES;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
inode_lock(inode);
if (f2fs_is_atomic_file(inode))
drop_inmem_pages(inode);
if (f2fs_is_volatile_file(inode)) {
clear_inode_flag(inode, FI_VOLATILE_FILE);
stat_dec_volatile_write(inode);
ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true);
}
inode_unlock(inode);
mnt_drop_write_file(filp);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
return ret;
}
static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct super_block *sb = sbi->sb;
__u32 in;
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (get_user(in, (__u32 __user *)arg))
return -EFAULT;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
switch (in) {
case F2FS_GOING_DOWN_FULLSYNC:
sb = freeze_bdev(sb->s_bdev);
if (sb && !IS_ERR(sb)) {
f2fs_stop_checkpoint(sbi, false);
thaw_bdev(sb->s_bdev, sb);
}
break;
case F2FS_GOING_DOWN_METASYNC:
/* do checkpoint only */
f2fs_sync_fs(sb, 1);
f2fs_stop_checkpoint(sbi, false);
break;
case F2FS_GOING_DOWN_NOSYNC:
f2fs_stop_checkpoint(sbi, false);
break;
case F2FS_GOING_DOWN_METAFLUSH:
sync_meta_pages(sbi, META, LONG_MAX, FS_META_IO);
f2fs_stop_checkpoint(sbi, false);
break;
default:
ret = -EINVAL;
goto out;
}
f2fs_update_time(sbi, REQ_TIME);
out:
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_fitrim(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct super_block *sb = inode->i_sb;
struct request_queue *q = bdev_get_queue(sb->s_bdev);
struct fstrim_range range;
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (!blk_queue_discard(q))
return -EOPNOTSUPP;
if (copy_from_user(&range, (struct fstrim_range __user *)arg,
sizeof(range)))
return -EFAULT;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
range.minlen = max((unsigned int)range.minlen,
q->limits.discard_granularity);
ret = f2fs_trim_fs(F2FS_SB(sb), &range);
mnt_drop_write_file(filp);
if (ret < 0)
return ret;
if (copy_to_user((struct fstrim_range __user *)arg, &range,
sizeof(range)))
return -EFAULT;
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
return 0;
}
static bool uuid_is_nonzero(__u8 u[16])
{
int i;
for (i = 0; i < 16; i++)
if (u[i])
return true;
return false;
}
static int f2fs_ioc_set_encryption_policy(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
if (!f2fs_sb_has_crypto(inode->i_sb))
return -EOPNOTSUPP;
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
return fscrypt_ioctl_set_policy(filp, (const void __user *)arg);
}
static int f2fs_ioc_get_encryption_policy(struct file *filp, unsigned long arg)
{
if (!f2fs_sb_has_crypto(file_inode(filp)->i_sb))
return -EOPNOTSUPP;
return fscrypt_ioctl_get_policy(filp, (void __user *)arg);
}
static int f2fs_ioc_get_encryption_pwsalt(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
int err;
if (!f2fs_sb_has_crypto(inode->i_sb))
return -EOPNOTSUPP;
if (uuid_is_nonzero(sbi->raw_super->encrypt_pw_salt))
goto got_it;
err = mnt_want_write_file(filp);
if (err)
return err;
/* update superblock with uuid */
generate_random_uuid(sbi->raw_super->encrypt_pw_salt);
err = f2fs_commit_super(sbi, false);
if (err) {
/* undo new data */
memset(sbi->raw_super->encrypt_pw_salt, 0, 16);
mnt_drop_write_file(filp);
return err;
}
mnt_drop_write_file(filp);
got_it:
if (copy_to_user((__u8 __user *)arg, sbi->raw_super->encrypt_pw_salt,
16))
return -EFAULT;
return 0;
}
static int f2fs_ioc_gc(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
__u32 sync;
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (get_user(sync, (__u32 __user *)arg))
return -EFAULT;
if (f2fs_readonly(sbi->sb))
return -EROFS;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
if (!sync) {
if (!mutex_trylock(&sbi->gc_mutex)) {
ret = -EBUSY;
goto out;
}
} else {
mutex_lock(&sbi->gc_mutex);
}
ret = f2fs_gc(sbi, sync, true, NULL_SEGNO);
out:
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_gc_range(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct f2fs_gc_range range;
u64 end;
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (copy_from_user(&range, (struct f2fs_gc_range __user *)arg,
sizeof(range)))
return -EFAULT;
if (f2fs_readonly(sbi->sb))
return -EROFS;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
end = range.start + range.len;
if (range.start < MAIN_BLKADDR(sbi) || end >= MAX_BLKADDR(sbi))
return -EINVAL;
do_more:
if (!range.sync) {
if (!mutex_trylock(&sbi->gc_mutex)) {
ret = -EBUSY;
goto out;
}
} else {
mutex_lock(&sbi->gc_mutex);
}
ret = f2fs_gc(sbi, range.sync, true, GET_SEGNO(sbi, range.start));
range.start += sbi->blocks_per_seg;
if (range.start <= end)
goto do_more;
out:
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (f2fs_readonly(sbi->sb))
return -EROFS;
ret = mnt_want_write_file(filp);
if (ret)
return ret;
ret = f2fs_sync_fs(sbi->sb, 1);
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
struct file *filp,
struct f2fs_defragment *range)
{
struct inode *inode = file_inode(filp);
struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
struct extent_info ei = {0,0,0};
pgoff_t pg_start, pg_end;
unsigned int blk_per_seg = sbi->blocks_per_seg;
unsigned int total = 0, sec_num;
block_t blk_end = 0;
bool fragmented = false;
int err;
/* if in-place-update policy is enabled, don't waste time here */
if (need_inplace_update_policy(inode, NULL))
return -EINVAL;
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
pg_start = range->start >> PAGE_SHIFT;
pg_end = (range->start + range->len) >> PAGE_SHIFT;
f2fs_balance_fs(sbi, true);
inode_lock(inode);
/* writeback all dirty pages in the range */
err = filemap_write_and_wait_range(inode->i_mapping, range->start,
range->start + range->len - 1);
if (err)
goto out;
/*
* lookup mapping info in extent cache, skip defragmenting if physical
* block addresses are continuous.
*/
if (f2fs_lookup_extent_cache(inode, pg_start, &ei)) {
if (ei.fofs + ei.len >= pg_end)
goto out;
}
map.m_lblk = pg_start;
/*
* lookup mapping info in dnode page cache, skip defragmenting if all
* physical block addresses are continuous even if there are hole(s)
* in logical blocks.
*/
while (map.m_lblk < pg_end) {
map.m_len = pg_end - map.m_lblk;
err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT);
if (err)
goto out;
if (!(map.m_flags & F2FS_MAP_FLAGS)) {
map.m_lblk++;
continue;
}
if (blk_end && blk_end != map.m_pblk) {
fragmented = true;
break;
}
blk_end = map.m_pblk + map.m_len;
map.m_lblk += map.m_len;
}
if (!fragmented)
goto out;
map.m_lblk = pg_start;
map.m_len = pg_end - pg_start;
sec_num = (map.m_len + BLKS_PER_SEC(sbi) - 1) / BLKS_PER_SEC(sbi);
/*
* make sure there are enough free section for LFS allocation, this can
* avoid defragment running in SSR mode when free section are allocated
* intensively
*/
if (has_not_enough_free_secs(sbi, 0, sec_num)) {
err = -EAGAIN;
goto out;
}
while (map.m_lblk < pg_end) {
pgoff_t idx;
int cnt = 0;
do_map:
map.m_len = pg_end - map.m_lblk;
err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT);
if (err)
goto clear_out;
if (!(map.m_flags & F2FS_MAP_FLAGS)) {
map.m_lblk++;
continue;
}
set_inode_flag(inode, FI_DO_DEFRAG);
idx = map.m_lblk;
while (idx < map.m_lblk + map.m_len && cnt < blk_per_seg) {
struct page *page;
page = get_lock_data_page(inode, idx, true);
if (IS_ERR(page)) {
err = PTR_ERR(page);
goto clear_out;
}
set_page_dirty(page);
f2fs_put_page(page, 1);
idx++;
cnt++;
total++;
}
map.m_lblk = idx;
if (idx < pg_end && cnt < blk_per_seg)
goto do_map;
clear_inode_flag(inode, FI_DO_DEFRAG);
err = filemap_fdatawrite(inode->i_mapping);
if (err)
goto out;
}
clear_out:
clear_inode_flag(inode, FI_DO_DEFRAG);
out:
inode_unlock(inode);
if (!err)
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
range->len = (u64)total << PAGE_SHIFT;
return err;
}
static int f2fs_ioc_defragment(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct f2fs_defragment range;
int err;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (!S_ISREG(inode->i_mode) || f2fs_is_atomic_file(inode))
return -EINVAL;
if (f2fs_readonly(sbi->sb))
return -EROFS;
if (copy_from_user(&range, (struct f2fs_defragment __user *)arg,
sizeof(range)))
return -EFAULT;
/* verify alignment of offset & size */
if (range.start & (F2FS_BLKSIZE - 1) || range.len & (F2FS_BLKSIZE - 1))
return -EINVAL;
if (unlikely((range.start + range.len) >> PAGE_SHIFT >
sbi->max_file_blocks))
return -EINVAL;
err = mnt_want_write_file(filp);
if (err)
return err;
err = f2fs_defragment_range(sbi, filp, &range);
mnt_drop_write_file(filp);
f2fs_update_time(sbi, REQ_TIME);
if (err < 0)
return err;
if (copy_to_user((struct f2fs_defragment __user *)arg, &range,
sizeof(range)))
return -EFAULT;
return 0;
}
static int f2fs_move_file_range(struct file *file_in, loff_t pos_in,
struct file *file_out, loff_t pos_out, size_t len)
{
struct inode *src = file_inode(file_in);
struct inode *dst = file_inode(file_out);
struct f2fs_sb_info *sbi = F2FS_I_SB(src);
size_t olen = len, dst_max_i_size = 0;
size_t dst_osize;
int ret;
if (file_in->f_path.mnt != file_out->f_path.mnt ||
src->i_sb != dst->i_sb)
return -EXDEV;
if (unlikely(f2fs_readonly(src->i_sb)))
return -EROFS;
if (!S_ISREG(src->i_mode) || !S_ISREG(dst->i_mode))
return -EINVAL;
if (f2fs_encrypted_inode(src) || f2fs_encrypted_inode(dst))
return -EOPNOTSUPP;
if (src == dst) {
if (pos_in == pos_out)
return 0;
if (pos_out > pos_in && pos_out < pos_in + len)
return -EINVAL;
}
inode_lock(src);
down_write(&F2FS_I(src)->dio_rwsem[WRITE]);
if (src != dst) {
ret = -EBUSY;
if (!inode_trylock(dst))
goto out;
if (!down_write_trylock(&F2FS_I(dst)->dio_rwsem[WRITE])) {
inode_unlock(dst);
goto out;
}
}
ret = -EINVAL;
if (pos_in + len > src->i_size || pos_in + len < pos_in)
goto out_unlock;
if (len == 0)
olen = len = src->i_size - pos_in;
if (pos_in + len == src->i_size)
len = ALIGN(src->i_size, F2FS_BLKSIZE) - pos_in;
if (len == 0) {
ret = 0;
goto out_unlock;
}
dst_osize = dst->i_size;
if (pos_out + olen > dst->i_size)
dst_max_i_size = pos_out + olen;
/* verify the end result is block aligned */
if (!IS_ALIGNED(pos_in, F2FS_BLKSIZE) ||
!IS_ALIGNED(pos_in + len, F2FS_BLKSIZE) ||
!IS_ALIGNED(pos_out, F2FS_BLKSIZE))
goto out_unlock;
ret = f2fs_convert_inline_inode(src);
if (ret)
goto out_unlock;
ret = f2fs_convert_inline_inode(dst);
if (ret)
goto out_unlock;
/* write out all dirty pages from offset */
ret = filemap_write_and_wait_range(src->i_mapping,
pos_in, pos_in + len);
if (ret)
goto out_unlock;
ret = filemap_write_and_wait_range(dst->i_mapping,
pos_out, pos_out + len);
if (ret)
goto out_unlock;
f2fs_balance_fs(sbi, true);
f2fs_lock_op(sbi);
ret = __exchange_data_block(src, dst, pos_in >> F2FS_BLKSIZE_BITS,
pos_out >> F2FS_BLKSIZE_BITS,
len >> F2FS_BLKSIZE_BITS, false);
if (!ret) {
if (dst_max_i_size)
f2fs_i_size_write(dst, dst_max_i_size);
else if (dst_osize != dst->i_size)
f2fs_i_size_write(dst, dst_osize);
}
f2fs_unlock_op(sbi);
out_unlock:
if (src != dst) {
up_write(&F2FS_I(dst)->dio_rwsem[WRITE]);
inode_unlock(dst);
}
out:
up_write(&F2FS_I(src)->dio_rwsem[WRITE]);
inode_unlock(src);
return ret;
}
static int f2fs_ioc_move_range(struct file *filp, unsigned long arg)
{
struct f2fs_move_range range;
struct fd dst;
int err;
if (!(filp->f_mode & FMODE_READ) ||
!(filp->f_mode & FMODE_WRITE))
return -EBADF;
if (copy_from_user(&range, (struct f2fs_move_range __user *)arg,
sizeof(range)))
return -EFAULT;
dst = fdget(range.dst_fd);
if (!dst.file)
return -EBADF;
if (!(dst.file->f_mode & FMODE_WRITE)) {
err = -EBADF;
goto err_out;
}
err = mnt_want_write_file(filp);
if (err)
goto err_out;
err = f2fs_move_file_range(filp, range.pos_in, dst.file,
range.pos_out, range.len);
mnt_drop_write_file(filp);
if (err)
goto err_out;
if (copy_to_user((struct f2fs_move_range __user *)arg,
&range, sizeof(range)))
err = -EFAULT;
err_out:
fdput(dst);
return err;
}
static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct sit_info *sm = SIT_I(sbi);
unsigned int start_segno = 0, end_segno = 0;
unsigned int dev_start_segno = 0, dev_end_segno = 0;
struct f2fs_flush_device range;
int ret;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (f2fs_readonly(sbi->sb))
return -EROFS;
if (copy_from_user(&range, (struct f2fs_flush_device __user *)arg,
sizeof(range)))
return -EFAULT;
if (sbi->s_ndevs <= 1 || sbi->s_ndevs - 1 <= range.dev_num ||
sbi->segs_per_sec != 1) {
f2fs_msg(sbi->sb, KERN_WARNING,
"Can't flush %u in %d for segs_per_sec %u != 1\n",
range.dev_num, sbi->s_ndevs,
sbi->segs_per_sec);
return -EINVAL;
}
ret = mnt_want_write_file(filp);
if (ret)
return ret;
if (range.dev_num != 0)
dev_start_segno = GET_SEGNO(sbi, FDEV(range.dev_num).start_blk);
dev_end_segno = GET_SEGNO(sbi, FDEV(range.dev_num).end_blk);
start_segno = sm->last_victim[FLUSH_DEVICE];
if (start_segno < dev_start_segno || start_segno >= dev_end_segno)
start_segno = dev_start_segno;
end_segno = min(start_segno + range.segments, dev_end_segno);
while (start_segno < end_segno) {
if (!mutex_trylock(&sbi->gc_mutex)) {
ret = -EBUSY;
goto out;
}
sm->last_victim[GC_CB] = end_segno + 1;
sm->last_victim[GC_GREEDY] = end_segno + 1;
sm->last_victim[ALLOC_NEXT] = end_segno + 1;
ret = f2fs_gc(sbi, true, true, start_segno);
if (ret == -EAGAIN)
ret = 0;
else if (ret < 0)
break;
start_segno++;
}
out:
mnt_drop_write_file(filp);
return ret;
}
static int f2fs_ioc_get_features(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
u32 sb_feature = le32_to_cpu(F2FS_I_SB(inode)->raw_super->feature);
/* Must validate to set it with SQLite behavior in Android. */
sb_feature |= F2FS_FEATURE_ATOMIC_WRITE;
return put_user(sb_feature, (u32 __user *)arg);
}
#ifdef CONFIG_QUOTA
static int f2fs_ioc_setproject(struct file *filp, __u32 projid)
{
struct inode *inode = file_inode(filp);
struct f2fs_inode_info *fi = F2FS_I(inode);
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct super_block *sb = sbi->sb;
struct dquot *transfer_to[MAXQUOTAS] = {};
struct page *ipage;
kprojid_t kprojid;
int err;
if (!f2fs_sb_has_project_quota(sb)) {
if (projid != F2FS_DEF_PROJID)
return -EOPNOTSUPP;
else
return 0;
}
if (!f2fs_has_extra_attr(inode))
return -EOPNOTSUPP;
kprojid = make_kprojid(&init_user_ns, (projid_t)projid);
if (projid_eq(kprojid, F2FS_I(inode)->i_projid))
return 0;
err = mnt_want_write_file(filp);
if (err)
return err;
err = -EPERM;
inode_lock(inode);
/* Is it quota file? Do not allow user to mess with it */
if (IS_NOQUOTA(inode))
goto out_unlock;
ipage = get_node_page(sbi, inode->i_ino);
if (IS_ERR(ipage)) {
err = PTR_ERR(ipage);
goto out_unlock;
}
if (!F2FS_FITS_IN_INODE(F2FS_INODE(ipage), fi->i_extra_isize,
i_projid)) {
err = -EOVERFLOW;
f2fs_put_page(ipage, 1);
goto out_unlock;
}
f2fs_put_page(ipage, 1);
dquot_initialize(inode);
transfer_to[PRJQUOTA] = dqget(sb, make_kqid_projid(kprojid));
if (!IS_ERR(transfer_to[PRJQUOTA])) {
err = __dquot_transfer(inode, transfer_to);
dqput(transfer_to[PRJQUOTA]);
if (err)
goto out_dirty;
}
F2FS_I(inode)->i_projid = kprojid;
inode->i_ctime = current_time(inode);
out_dirty:
f2fs_mark_inode_dirty_sync(inode, true);
out_unlock:
inode_unlock(inode);
mnt_drop_write_file(filp);
return err;
}
#else
static int f2fs_ioc_setproject(struct file *filp, __u32 projid)
{
if (projid != F2FS_DEF_PROJID)
return -EOPNOTSUPP;
return 0;
}
#endif
/* Transfer internal flags to xflags */
static inline __u32 f2fs_iflags_to_xflags(unsigned long iflags)
{
__u32 xflags = 0;
if (iflags & FS_SYNC_FL)
xflags |= FS_XFLAG_SYNC;
if (iflags & FS_IMMUTABLE_FL)
xflags |= FS_XFLAG_IMMUTABLE;
if (iflags & FS_APPEND_FL)
xflags |= FS_XFLAG_APPEND;
if (iflags & FS_NODUMP_FL)
xflags |= FS_XFLAG_NODUMP;
if (iflags & FS_NOATIME_FL)
xflags |= FS_XFLAG_NOATIME;
if (iflags & FS_PROJINHERIT_FL)
xflags |= FS_XFLAG_PROJINHERIT;
return xflags;
}
#define F2FS_SUPPORTED_FS_XFLAGS (FS_XFLAG_SYNC | FS_XFLAG_IMMUTABLE | \
FS_XFLAG_APPEND | FS_XFLAG_NODUMP | \
FS_XFLAG_NOATIME | FS_XFLAG_PROJINHERIT)
/* Flags we can manipulate with through EXT4_IOC_FSSETXATTR */
#define F2FS_FL_XFLAG_VISIBLE (FS_SYNC_FL | \
FS_IMMUTABLE_FL | \
FS_APPEND_FL | \
FS_NODUMP_FL | \
FS_NOATIME_FL | \
FS_PROJINHERIT_FL)
/* Transfer xflags flags to internal */
static inline unsigned long f2fs_xflags_to_iflags(__u32 xflags)
{
unsigned long iflags = 0;
if (xflags & FS_XFLAG_SYNC)
iflags |= FS_SYNC_FL;
if (xflags & FS_XFLAG_IMMUTABLE)
iflags |= FS_IMMUTABLE_FL;
if (xflags & FS_XFLAG_APPEND)
iflags |= FS_APPEND_FL;
if (xflags & FS_XFLAG_NODUMP)
iflags |= FS_NODUMP_FL;
if (xflags & FS_XFLAG_NOATIME)
iflags |= FS_NOATIME_FL;
if (xflags & FS_XFLAG_PROJINHERIT)
iflags |= FS_PROJINHERIT_FL;
return iflags;
}
static int f2fs_ioc_fsgetxattr(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_inode_info *fi = F2FS_I(inode);
struct fsxattr fa;
memset(&fa, 0, sizeof(struct fsxattr));
fa.fsx_xflags = f2fs_iflags_to_xflags(fi->i_flags &
(FS_FL_USER_VISIBLE | FS_PROJINHERIT_FL));
if (f2fs_sb_has_project_quota(inode->i_sb))
fa.fsx_projid = (__u32)from_kprojid(&init_user_ns,
fi->i_projid);
if (copy_to_user((struct fsxattr __user *)arg, &fa, sizeof(fa)))
return -EFAULT;
return 0;
}
static int f2fs_ioc_fssetxattr(struct file *filp, unsigned long arg)
{
struct inode *inode = file_inode(filp);
struct f2fs_inode_info *fi = F2FS_I(inode);
struct fsxattr fa;
unsigned int flags;
int err;
if (copy_from_user(&fa, (struct fsxattr __user *)arg, sizeof(fa)))
return -EFAULT;
/* Make sure caller has proper permission */
if (!inode_owner_or_capable(inode))
return -EACCES;
if (fa.fsx_xflags & ~F2FS_SUPPORTED_FS_XFLAGS)
return -EOPNOTSUPP;
flags = f2fs_xflags_to_iflags(fa.fsx_xflags);
if (f2fs_mask_flags(inode->i_mode, flags) != flags)
return -EOPNOTSUPP;
err = mnt_want_write_file(filp);
if (err)
return err;
inode_lock(inode);
flags = (fi->i_flags & ~F2FS_FL_XFLAG_VISIBLE) |
(flags & F2FS_FL_XFLAG_VISIBLE);
err = __f2fs_ioc_setflags(inode, flags);
inode_unlock(inode);
mnt_drop_write_file(filp);
if (err)
return err;
err = f2fs_ioc_setproject(filp, fa.fsx_projid);
if (err)
return err;
return 0;
}
long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
if (unlikely(f2fs_cp_error(F2FS_I_SB(file_inode(filp)))))
return -EIO;
switch (cmd) {
case F2FS_IOC_GETFLAGS:
return f2fs_ioc_getflags(filp, arg);
case F2FS_IOC_SETFLAGS:
return f2fs_ioc_setflags(filp, arg);
case F2FS_IOC_GETVERSION:
return f2fs_ioc_getversion(filp, arg);
case F2FS_IOC_START_ATOMIC_WRITE:
return f2fs_ioc_start_atomic_write(filp);
case F2FS_IOC_COMMIT_ATOMIC_WRITE:
return f2fs_ioc_commit_atomic_write(filp);
case F2FS_IOC_START_VOLATILE_WRITE:
return f2fs_ioc_start_volatile_write(filp);
case F2FS_IOC_RELEASE_VOLATILE_WRITE:
return f2fs_ioc_release_volatile_write(filp);
case F2FS_IOC_ABORT_VOLATILE_WRITE:
return f2fs_ioc_abort_volatile_write(filp);
case F2FS_IOC_SHUTDOWN:
return f2fs_ioc_shutdown(filp, arg);
case FITRIM:
return f2fs_ioc_fitrim(filp, arg);
case F2FS_IOC_SET_ENCRYPTION_POLICY:
return f2fs_ioc_set_encryption_policy(filp, arg);
case F2FS_IOC_GET_ENCRYPTION_POLICY:
return f2fs_ioc_get_encryption_policy(filp, arg);
case F2FS_IOC_GET_ENCRYPTION_PWSALT:
return f2fs_ioc_get_encryption_pwsalt(filp, arg);
case F2FS_IOC_GARBAGE_COLLECT:
return f2fs_ioc_gc(filp, arg);
case F2FS_IOC_GARBAGE_COLLECT_RANGE:
return f2fs_ioc_gc_range(filp, arg);
case F2FS_IOC_WRITE_CHECKPOINT:
return f2fs_ioc_write_checkpoint(filp, arg);
case F2FS_IOC_DEFRAGMENT:
return f2fs_ioc_defragment(filp, arg);
case F2FS_IOC_MOVE_RANGE:
return f2fs_ioc_move_range(filp, arg);
case F2FS_IOC_FLUSH_DEVICE:
return f2fs_ioc_flush_device(filp, arg);
case F2FS_IOC_GET_FEATURES:
return f2fs_ioc_get_features(filp, arg);
case F2FS_IOC_FSGETXATTR:
return f2fs_ioc_fsgetxattr(filp, arg);
case F2FS_IOC_FSSETXATTR:
return f2fs_ioc_fssetxattr(filp, arg);
default:
return -ENOTTY;
}
}
static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file_inode(file);
struct blk_plug plug;
ssize_t ret;
if (unlikely(f2fs_cp_error(F2FS_I_SB(inode))))
return -EIO;
inode_lock(inode);
ret = generic_write_checks(iocb, from);
if (ret > 0) {
int err;
if (iov_iter_fault_in_readable(from, iov_iter_count(from)))
set_inode_flag(inode, FI_NO_PREALLOC);
err = f2fs_preallocate_blocks(iocb, from);
if (err) {
clear_inode_flag(inode, FI_NO_PREALLOC);
inode_unlock(inode);
return err;
}
blk_start_plug(&plug);
ret = __generic_file_write_iter(iocb, from);
blk_finish_plug(&plug);
clear_inode_flag(inode, FI_NO_PREALLOC);
if (ret > 0)
f2fs_update_iostat(F2FS_I_SB(inode), APP_WRITE_IO, ret);
}
inode_unlock(inode);
if (ret > 0)
ret = generic_write_sync(iocb, ret);
return ret;
}
#ifdef CONFIG_COMPAT
long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
switch (cmd) {
case F2FS_IOC32_GETFLAGS:
cmd = F2FS_IOC_GETFLAGS;
break;
case F2FS_IOC32_SETFLAGS:
cmd = F2FS_IOC_SETFLAGS;
break;
case F2FS_IOC32_GETVERSION:
cmd = F2FS_IOC_GETVERSION;
break;
case F2FS_IOC_START_ATOMIC_WRITE:
case F2FS_IOC_COMMIT_ATOMIC_WRITE:
case F2FS_IOC_START_VOLATILE_WRITE:
case F2FS_IOC_RELEASE_VOLATILE_WRITE:
case F2FS_IOC_ABORT_VOLATILE_WRITE:
case F2FS_IOC_SHUTDOWN:
case F2FS_IOC_SET_ENCRYPTION_POLICY:
case F2FS_IOC_GET_ENCRYPTION_PWSALT:
case F2FS_IOC_GET_ENCRYPTION_POLICY:
case F2FS_IOC_GARBAGE_COLLECT:
case F2FS_IOC_GARBAGE_COLLECT_RANGE:
case F2FS_IOC_WRITE_CHECKPOINT:
case F2FS_IOC_DEFRAGMENT:
case F2FS_IOC_MOVE_RANGE:
case F2FS_IOC_FLUSH_DEVICE:
case F2FS_IOC_GET_FEATURES:
case F2FS_IOC_FSGETXATTR:
case F2FS_IOC_FSSETXATTR:
break;
default:
return -ENOIOCTLCMD;
}
return f2fs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
}
#endif
const struct file_operations f2fs_file_operations = {
.llseek = f2fs_llseek,
.read_iter = generic_file_read_iter,
.write_iter = f2fs_file_write_iter,
.open = f2fs_file_open,
.release = f2fs_release_file,
.mmap = f2fs_file_mmap,
.flush = f2fs_file_flush,
.fsync = f2fs_sync_file,
.fallocate = f2fs_fallocate,
.unlocked_ioctl = f2fs_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = f2fs_compat_ioctl,
#endif
.splice_read = generic_file_splice_read,
.splice_write = iter_file_splice_write,
};