kernel-ark/fs/nilfs2/btnode.c

298 lines
7.7 KiB
C
Raw Normal View History

/*
* btnode.c - NILFS B-tree node cache
*
* Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*
* This file was originally written by Seiji Kihara <kihara@osrg.net>
* and fully revised by Ryusuke Konishi <ryusuke@osrg.net> for
* stabilization and simplification.
*
*/
#include <linux/types.h>
#include <linux/buffer_head.h>
#include <linux/mm.h>
#include <linux/backing-dev.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/gfp.h>
#include "nilfs.h"
#include "mdt.h"
#include "dat.h"
#include "page.h"
#include "btnode.h"
void nilfs_btnode_cache_clear(struct address_space *btnc)
{
invalidate_mapping_pages(btnc, 0, -1);
truncate_inode_pages(btnc, 0);
}
struct buffer_head *
nilfs_btnode_create_block(struct address_space *btnc, __u64 blocknr)
{
struct inode *inode = NILFS_BTNC_I(btnc);
struct buffer_head *bh;
bh = nilfs_grab_buffer(inode, btnc, blocknr, 1 << BH_NILFS_Node);
if (unlikely(!bh))
return NULL;
if (unlikely(buffer_mapped(bh) || buffer_uptodate(bh) ||
buffer_dirty(bh))) {
brelse(bh);
BUG();
}
memset(bh->b_data, 0, 1 << inode->i_blkbits);
bh->b_bdev = inode->i_sb->s_bdev;
bh->b_blocknr = blocknr;
set_buffer_mapped(bh);
set_buffer_uptodate(bh);
unlock_page(bh->b_page);
page_cache_release(bh->b_page);
return bh;
}
int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr,
sector_t pblocknr, int mode,
struct buffer_head **pbh, sector_t *submit_ptr)
{
struct buffer_head *bh;
struct inode *inode = NILFS_BTNC_I(btnc);
struct page *page;
int err;
bh = nilfs_grab_buffer(inode, btnc, blocknr, 1 << BH_NILFS_Node);
if (unlikely(!bh))
return -ENOMEM;
err = -EEXIST; /* internal code */
page = bh->b_page;
if (buffer_uptodate(bh) || buffer_dirty(bh))
goto found;
if (pblocknr == 0) {
pblocknr = blocknr;
if (inode->i_ino != NILFS_DAT_INO) {
struct the_nilfs *nilfs = inode->i_sb->s_fs_info;
/* blocknr is a virtual block number */
err = nilfs_dat_translate(nilfs->ns_dat, blocknr,
&pblocknr);
if (unlikely(err)) {
brelse(bh);
goto out_locked;
}
}
}
if (mode == READA) {
if (pblocknr != *submit_ptr + 1 || !trylock_buffer(bh)) {
err = -EBUSY; /* internal code */
brelse(bh);
goto out_locked;
}
} else { /* mode == READ */
lock_buffer(bh);
}
if (buffer_uptodate(bh)) {
unlock_buffer(bh);
err = -EEXIST; /* internal code */
goto found;
}
set_buffer_mapped(bh);
bh->b_bdev = inode->i_sb->s_bdev;
bh->b_blocknr = pblocknr; /* set block address for read */
bh->b_end_io = end_buffer_read_sync;
get_bh(bh);
submit_bh(mode, bh);
bh->b_blocknr = blocknr; /* set back to the given block address */
*submit_ptr = pblocknr;
err = 0;
found:
*pbh = bh;
out_locked:
unlock_page(page);
page_cache_release(page);
return err;
}
/**
* nilfs_btnode_delete - delete B-tree node buffer
* @bh: buffer to be deleted
*
* nilfs_btnode_delete() invalidates the specified buffer and delete the page
* including the buffer if the page gets unbusy.
*/
void nilfs_btnode_delete(struct buffer_head *bh)
{
struct address_space *mapping;
struct page *page = bh->b_page;
pgoff_t index = page_index(page);
int still_dirty;
page_cache_get(page);
lock_page(page);
wait_on_page_writeback(page);
nilfs_forget_buffer(bh);
still_dirty = PageDirty(page);
mapping = page->mapping;
unlock_page(page);
page_cache_release(page);
if (!still_dirty && mapping)
invalidate_inode_pages2_range(mapping, index, index);
}
/**
* nilfs_btnode_prepare_change_key
* prepare to move contents of the block for old key to one of new key.
* the old buffer will not be removed, but might be reused for new buffer.
* it might return -ENOMEM because of memory allocation errors,
* and might return -EIO because of disk read errors.
*/
int nilfs_btnode_prepare_change_key(struct address_space *btnc,
struct nilfs_btnode_chkey_ctxt *ctxt)
{
struct buffer_head *obh, *nbh;
struct inode *inode = NILFS_BTNC_I(btnc);
__u64 oldkey = ctxt->oldkey, newkey = ctxt->newkey;
int err;
if (oldkey == newkey)
return 0;
obh = ctxt->bh;
ctxt->newbh = NULL;
if (inode->i_blkbits == PAGE_CACHE_SHIFT) {
lock_page(obh->b_page);
/*
* We cannot call radix_tree_preload for the kernels older
* than 2.6.23, because it is not exported for modules.
*/
nilfs2: fix preempt count underflow in nilfs_btnode_prepare_change_key This will fix the following preempt count underflow reported from users with the title "[NILFS users] segctord problem" (Message-ID: <949415.6494.qm@web58808.mail.re1.yahoo.com> and Message-ID: <debc30fc0908270825v747c1734xa59126623cfd5b05@mail.gmail.com>): WARNING: at kernel/sched.c:4890 sub_preempt_count+0x95/0xa0() Hardware name: HP Compaq 6530b (KR980UT#ABC) Modules linked in: bridge stp llc bnep rfcomm l2cap xfs exportfs nilfs2 cowloop loop vboxnetadp vboxnetflt vboxdrv btusb bluetooth uvcvideo videodev v4l1_compat v4l2_compat_ioctl32 arc4 snd_hda_codec_analog ecb iwlagn iwlcore rfkill lib80211 mac80211 snd_hda_intel snd_hda_codec ehci_hcd uhci_hcd usbcore snd_hwdep snd_pcm tg3 cfg80211 psmouse snd_timer joydev libphy ohci1394 snd_page_alloc hp_accel lis3lv02d ieee1394 led_class i915 drm i2c_algo_bit video backlight output i2c_core dm_crypt dm_mod Pid: 4197, comm: segctord Not tainted 2.6.30-gentoo-r4-64 #7 Call Trace: [<ffffffff8023fa05>] ? sub_preempt_count+0x95/0xa0 [<ffffffff802470f8>] warn_slowpath_common+0x78/0xd0 [<ffffffff8024715f>] warn_slowpath_null+0xf/0x20 [<ffffffff8023fa05>] sub_preempt_count+0x95/0xa0 [<ffffffffa04ce4db>] nilfs_btnode_prepare_change_key+0x11b/0x190 [nilfs2] [<ffffffffa04d01ad>] nilfs_btree_assign_p+0x19d/0x1e0 [nilfs2] [<ffffffffa04d10ad>] nilfs_btree_assign+0xbd/0x130 [nilfs2] [<ffffffffa04cead7>] nilfs_bmap_assign+0x47/0x70 [nilfs2] [<ffffffffa04d9bc6>] nilfs_segctor_do_construct+0x956/0x20f0 [nilfs2] [<ffffffff805ac8e2>] ? _spin_unlock_irqrestore+0x12/0x40 [<ffffffff803c06e0>] ? __up_write+0xe0/0x150 [<ffffffff80262959>] ? up_write+0x9/0x10 [<ffffffffa04ce9f3>] ? nilfs_bmap_test_and_clear_dirty+0x43/0x60 [nilfs2] [<ffffffffa04cd627>] ? nilfs_mdt_fetch_dirty+0x27/0x60 [nilfs2] [<ffffffffa04db5fc>] nilfs_segctor_construct+0x8c/0xd0 [nilfs2] [<ffffffffa04dc3dc>] nilfs_segctor_thread+0x15c/0x3a0 [nilfs2] [<ffffffffa04dbe20>] ? nilfs_construction_timeout+0x0/0x10 [nilfs2] [<ffffffff80252633>] ? add_timer+0x13/0x20 [<ffffffff802370da>] ? __wake_up_common+0x5a/0x90 [<ffffffff8025e960>] ? autoremove_wake_function+0x0/0x40 [<ffffffffa04dc280>] ? nilfs_segctor_thread+0x0/0x3a0 [nilfs2] [<ffffffffa04dc280>] ? nilfs_segctor_thread+0x0/0x3a0 [nilfs2] [<ffffffff8025e556>] kthread+0x56/0x90 [<ffffffff8020cdea>] child_rip+0xa/0x20 [<ffffffff8025e500>] ? kthread+0x0/0x90 [<ffffffff8020cde0>] ? child_rip+0x0/0x20 This problem was caused due to a missing radix_tree_preload() call in the retry path of nilfs_btnode_prepare_change_key() function. Reported-by: Eric A <eric225125@yahoo.com> Reported-by: Jerome Poulin <jeromepoulin@gmail.com> Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Tested-by: Jerome Poulin <jeromepoulin@gmail.com> Cc: stable@kernel.org
2009-08-29 19:21:41 +00:00
retry:
err = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
if (err)
goto failed_unlock;
/* BUG_ON(oldkey != obh->b_page->index); */
if (unlikely(oldkey != obh->b_page->index))
NILFS_PAGE_BUG(obh->b_page,
"invalid oldkey %lld (newkey=%lld)",
(unsigned long long)oldkey,
(unsigned long long)newkey);
spin_lock_irq(&btnc->tree_lock);
err = radix_tree_insert(&btnc->page_tree, newkey, obh->b_page);
spin_unlock_irq(&btnc->tree_lock);
/*
* Note: page->index will not change to newkey until
* nilfs_btnode_commit_change_key() will be called.
* To protect the page in intermediate state, the page lock
* is held.
*/
radix_tree_preload_end();
if (!err)
return 0;
else if (err != -EEXIST)
goto failed_unlock;
err = invalidate_inode_pages2_range(btnc, newkey, newkey);
if (!err)
goto retry;
/* fallback to copy mode */
unlock_page(obh->b_page);
}
nbh = nilfs_btnode_create_block(btnc, newkey);
if (!nbh)
return -ENOMEM;
BUG_ON(nbh == obh);
ctxt->newbh = nbh;
return 0;
failed_unlock:
unlock_page(obh->b_page);
return err;
}
/**
* nilfs_btnode_commit_change_key
* commit the change_key operation prepared by prepare_change_key().
*/
void nilfs_btnode_commit_change_key(struct address_space *btnc,
struct nilfs_btnode_chkey_ctxt *ctxt)
{
struct buffer_head *obh = ctxt->bh, *nbh = ctxt->newbh;
__u64 oldkey = ctxt->oldkey, newkey = ctxt->newkey;
struct page *opage;
if (oldkey == newkey)
return;
if (nbh == NULL) { /* blocksize == pagesize */
opage = obh->b_page;
if (unlikely(oldkey != opage->index))
NILFS_PAGE_BUG(opage,
"invalid oldkey %lld (newkey=%lld)",
(unsigned long long)oldkey,
(unsigned long long)newkey);
mark_buffer_dirty(obh);
spin_lock_irq(&btnc->tree_lock);
radix_tree_delete(&btnc->page_tree, oldkey);
radix_tree_tag_set(&btnc->page_tree, newkey,
PAGECACHE_TAG_DIRTY);
spin_unlock_irq(&btnc->tree_lock);
opage->index = obh->b_blocknr = newkey;
unlock_page(opage);
} else {
nilfs_copy_buffer(nbh, obh);
mark_buffer_dirty(nbh);
nbh->b_blocknr = newkey;
ctxt->bh = nbh;
nilfs_btnode_delete(obh); /* will decrement bh->b_count */
}
}
/**
* nilfs_btnode_abort_change_key
* abort the change_key operation prepared by prepare_change_key().
*/
void nilfs_btnode_abort_change_key(struct address_space *btnc,
struct nilfs_btnode_chkey_ctxt *ctxt)
{
struct buffer_head *nbh = ctxt->newbh;
__u64 oldkey = ctxt->oldkey, newkey = ctxt->newkey;
if (oldkey == newkey)
return;
if (nbh == NULL) { /* blocksize == pagesize */
spin_lock_irq(&btnc->tree_lock);
radix_tree_delete(&btnc->page_tree, newkey);
spin_unlock_irq(&btnc->tree_lock);
unlock_page(ctxt->bh->b_page);
} else
brelse(nbh);
}