From: "Theodore Ts'o" <tytso@mit.edu>
To: linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
Mingming Cao <cmm@us.ibm.com>, "Theodore Ts'o" <tytso@mit.edu>
Subject: [PATCH 13/42] ext4: Retry block allocation if we have free blocks left
Date: Thu, 9 Oct 2008 00:05:31 -0400 [thread overview]
Message-ID: <1223525160-9887-14-git-send-email-tytso@mit.edu> (raw)
In-Reply-To: <1223525160-9887-13-git-send-email-tytso@mit.edu>
From: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
When we truncate files, the meta-data blocks released are not reused
untill we commit the truncate transaction. That means delayed get_block
request will return ENOSPC even if we have free blocks left. Force a
journal commit and retry block allocation if we get ENOSPC with free
blocks left.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
fs/ext4/inode.c | 81 ++++++++++++++++++++++++++++++++++++++----------------
1 files changed, 57 insertions(+), 24 deletions(-)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index b567e71..f97b347 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1634,6 +1634,7 @@ struct mpage_da_data {
struct writeback_control *wbc;
int io_done;
long pages_written;
+ int retval;
};
/*
@@ -1820,6 +1821,24 @@ static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd,
return;
}
+static void ext4_print_free_blocks(struct inode *inode)
+{
+ struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+ printk(KERN_EMERG "Total free blocks count %lld\n",
+ ext4_count_free_blocks(inode->i_sb));
+ printk(KERN_EMERG "Free/Dirty block details\n");
+ printk(KERN_EMERG "free_blocks=%lld\n",
+ percpu_counter_sum(&sbi->s_freeblocks_counter));
+ printk(KERN_EMERG "dirty_blocks=%lld\n",
+ percpu_counter_sum(&sbi->s_dirtyblocks_counter));
+ printk(KERN_EMERG "Block reservation details\n");
+ printk(KERN_EMERG "i_reserved_data_blocks=%lu\n",
+ EXT4_I(inode)->i_reserved_data_blocks);
+ printk(KERN_EMERG "i_reserved_meta_blocks=%lu\n",
+ EXT4_I(inode)->i_reserved_meta_blocks);
+ return;
+}
+
/*
* mpage_da_map_blocks - go through given space
*
@@ -1834,7 +1853,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
int err = 0;
struct buffer_head new;
struct buffer_head *lbh = &mpd->lbh;
- sector_t next = lbh->b_blocknr;
+ sector_t next;
/*
* We consider only non-mapped and non-allocated blocks
@@ -1844,6 +1863,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
new.b_state = lbh->b_state;
new.b_blocknr = 0;
new.b_size = lbh->b_size;
+ next = lbh->b_blocknr;
/*
* If we didn't accumulate anything
* to write simply return
@@ -1860,6 +1880,13 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
*/
if (err == -EAGAIN)
return 0;
+
+ if (err == -ENOSPC &&
+ ext4_count_free_blocks(mpd->inode->i_sb)) {
+ mpd->retval = err;
+ return 0;
+ }
+
/*
* get block failure will cause us
* to loop in writepages. Because
@@ -1877,8 +1904,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
printk(KERN_EMERG "This should not happen.!! "
"Data will be lost\n");
if (err == -ENOSPC) {
- printk(KERN_CRIT "Total free blocks count %lld\n",
- ext4_count_free_blocks(mpd->inode->i_sb));
+ ext4_print_free_blocks(mpd->inode);
}
/* invlaidate all the pages */
ext4_da_block_invalidatepages(mpd, next,
@@ -2085,39 +2111,36 @@ static int __mpage_da_writepage(struct page *page,
*/
static int mpage_da_writepages(struct address_space *mapping,
struct writeback_control *wbc,
- get_block_t get_block)
+ struct mpage_da_data *mpd)
{
- struct mpage_da_data mpd;
long to_write;
int ret;
- if (!get_block)
+ if (!mpd->get_block)
return generic_writepages(mapping, wbc);
- mpd.wbc = wbc;
- mpd.inode = mapping->host;
- mpd.lbh.b_size = 0;
- mpd.lbh.b_state = 0;
- mpd.lbh.b_blocknr = 0;
- mpd.first_page = 0;
- mpd.next_page = 0;
- mpd.get_block = get_block;
- mpd.io_done = 0;
- mpd.pages_written = 0;
+ mpd->lbh.b_size = 0;
+ mpd->lbh.b_state = 0;
+ mpd->lbh.b_blocknr = 0;
+ mpd->first_page = 0;
+ mpd->next_page = 0;
+ mpd->io_done = 0;
+ mpd->pages_written = 0;
+ mpd->retval = 0;
to_write = wbc->nr_to_write;
- ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd);
+ ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, mpd);
/*
* Handle last extent of pages
*/
- if (!mpd.io_done && mpd.next_page != mpd.first_page) {
- if (mpage_da_map_blocks(&mpd) == 0)
- mpage_da_submit_io(&mpd);
+ if (!mpd->io_done && mpd->next_page != mpd->first_page) {
+ if (mpage_da_map_blocks(mpd) == 0)
+ mpage_da_submit_io(mpd);
}
- wbc->nr_to_write = to_write - mpd.pages_written;
+ wbc->nr_to_write = to_write - mpd->pages_written;
return ret;
}
@@ -2357,6 +2380,7 @@ static int ext4_da_writepages(struct address_space *mapping,
{
handle_t *handle = NULL;
loff_t range_start = 0;
+ struct mpage_da_data mpd;
struct inode *inode = mapping->host;
int needed_blocks, ret = 0, nr_to_writebump = 0;
long to_write, pages_skipped = 0;
@@ -2390,6 +2414,9 @@ static int ext4_da_writepages(struct address_space *mapping,
range_start = wbc->range_start;
pages_skipped = wbc->pages_skipped;
+ mpd.wbc = wbc;
+ mpd.inode = mapping->host;
+
restart_loop:
to_write = wbc->nr_to_write;
while (!ret && to_write > 0) {
@@ -2413,11 +2440,17 @@ restart_loop:
dump_stack();
goto out_writepages;
}
-
to_write -= wbc->nr_to_write;
- ret = mpage_da_writepages(mapping, wbc,
- ext4_da_get_block_write);
+
+ mpd.get_block = ext4_da_get_block_write;
+ ret = mpage_da_writepages(mapping, wbc, &mpd);
+
ext4_journal_stop(handle);
+
+ if (mpd.retval == -ENOSPC)
+ jbd2_journal_force_commit_nested(sbi->s_journal);
+
+ /* reset the retry count */
if (ret == MPAGE_DA_EXTENT_TAIL) {
/*
* got one extent now try with
--
1.5.6.1.205.ge2c7.dirty
next prev parent reply other threads:[~2008-10-09 4:15 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-09 4:05 [PATCH 0/42] Ext4 patches queued up for the 2.6.28 merge window Theodore Ts'o
2008-10-09 4:05 ` [PATCH 01/42] percpu counter: clean up percpu_counter_sum_and_set() Theodore Ts'o
2008-10-09 4:05 ` [PATCH 02/42] ext4: Add printk priority levels to clean up checkpatch warnings Theodore Ts'o
2008-10-09 4:05 ` [PATCH 03/42] ext4: Fix long long " Theodore Ts'o
2008-10-09 4:05 ` [PATCH 04/42] ext4: Fix whitespace checkpatch warnings/errors Theodore Ts'o
2008-10-09 4:05 ` [PATCH 05/42] ext4: invalidate pages if delalloc block allocation fails Theodore Ts'o
2008-10-09 4:05 ` [PATCH 06/42] ext4: Make sure all the block allocation paths reserve blocks Theodore Ts'o
2008-10-09 4:05 ` [PATCH 07/42] ext4: Retry block reservation Theodore Ts'o
2008-10-09 4:05 ` [PATCH 08/42] ext4: Add percpu dirty block accounting Theodore Ts'o
2008-10-09 4:05 ` [PATCH 09/42] ext4: Switch to non delalloc mode when we are low on free blocks count Theodore Ts'o
2008-10-09 4:05 ` [PATCH 10/42] ext4: Signed arithmetic fix Theodore Ts'o
2008-10-09 4:05 ` [PATCH 11/42] ext4: Fix ext4 nomballoc allocator for ENOSPC Theodore Ts'o
2008-10-09 4:05 ` [PATCH 12/42] ext4: Don't add the inode to journal handle until after the block is allocated Theodore Ts'o
2008-10-09 4:05 ` Theodore Ts'o [this message]
2008-10-09 4:05 ` [PATCH 14/42] ext4: truncate block allocated on a failed ext4_write_begin Theodore Ts'o
2008-10-09 4:05 ` [PATCH 15/42] ext4: Properly update i_disksize Theodore Ts'o
2008-10-09 4:05 ` [PATCH 16/42] ext4: Avoid printk floods in the face of directory corruption Theodore Ts'o
2008-10-09 4:05 ` [PATCH 17/42] Update flex_bg free blocks and free inodes counters when resizing Theodore Ts'o
2008-10-09 4:05 ` [PATCH 18/42] ext4: fix #11321: create /proc/ext4/*/stats more carefully Theodore Ts'o
2008-10-09 4:05 ` [PATCH 19/42] jbd2: clean up how the journal device name is printed Theodore Ts'o
2008-10-09 4:05 ` [PATCH 20/42] ext4: add missing unlock in ext4_check_descriptors() on error path Theodore Ts'o
2008-10-09 4:05 ` [PATCH 21/42] ext4: elevate write count for migrate ioctl Theodore Ts'o
2008-10-09 4:05 ` [PATCH 22/42] ext4: hook the ext3 migration interface to the EXT4_IOC_SETFLAGS ioctl Theodore Ts'o
2008-10-09 4:05 ` [PATCH 23/42] ext4: Renumber EXT4_IOC_MIGRATE Theodore Ts'o
2008-10-09 4:05 ` [PATCH 24/42] ext4: use percpu data structures for lg_prealloc_list Theodore Ts'o
2008-10-09 4:05 ` [PATCH 25/42] ext4/jbd2: Avoid WARN() messages when failing to write to the superblock Theodore Ts'o
2008-10-09 4:05 ` [PATCH 26/42] ext4: Don't use 'struct dentry' for internal lookups Theodore Ts'o
2008-10-09 4:05 ` [PATCH 27/42] ext4: move /proc setup and teardown out of mballoc.c Theodore Ts'o
2008-10-09 4:05 ` [PATCH 28/42] ext4: Combine proc file handling into a single set of functions Theodore Ts'o
2008-10-09 4:05 ` [PATCH 29/42] ext4: Use readahead when reading an inode from the inode table Theodore Ts'o
2008-10-09 4:05 ` [PATCH 30/42] ext4: Remove old legacy block allocator Theodore Ts'o
2008-10-09 4:05 ` [PATCH 31/42] ext4: fix initialization of UNINIT bitmap blocks Theodore Ts'o
2008-10-09 4:05 ` [PATCH 32/42] jbd2: abort instead of waiting for nonexistent transaction Theodore Ts'o
2008-10-09 4:05 ` [PATCH 33/42] ext4: Add debugging markers that can be used by systemtap Theodore Ts'o
2008-10-09 4:05 ` [PATCH 34/42] jbd2: Fix buffer head leak when writing the commit block Theodore Ts'o
2008-10-09 4:05 ` [PATCH 35/42] ext4: fix xattr deadlock Theodore Ts'o
2008-10-09 4:05 ` [PATCH 36/42] vfs: vfs-level fiemap interface Theodore Ts'o
2008-10-09 4:05 ` [PATCH 37/42] ocfs2: fiemap support Theodore Ts'o
2008-10-09 4:05 ` [PATCH 38/42] generic block based fiemap implementation Theodore Ts'o
2008-10-09 4:05 ` [PATCH 39/42] Hook ext4 to the vfs fiemap interface Theodore Ts'o
2008-10-09 4:05 ` [PATCH 40/42] Update ext4 MAINTAINERS file Theodore Ts'o
2008-10-09 4:05 ` [PATCH 41/42] ext4: Avoid double dirtying of super block in ext4_put_super() Theodore Ts'o
2008-10-09 4:06 ` [PATCH 42/42] ext4: Rename ext4dev to ext4 Theodore Ts'o
2008-10-11 22:04 ` Jeremy Fitzhardinge
2008-10-11 22:09 ` Eric Sandeen
2008-10-11 22:54 ` Jeremy Fitzhardinge
2008-10-11 22:58 ` Theodore Tso
2008-10-11 23:08 ` Grant Coady
2008-10-12 1:06 ` Eric Sandeen
2008-10-09 8:18 ` [PATCH 29/42] ext4: Use readahead when reading an inode from the inode table Aneesh Kumar K.V
2008-10-09 8:52 ` [PATCH 01/42] percpu counter: clean up percpu_counter_sum_and_set() Peter Zijlstra
2008-10-09 16:52 ` Theodore Tso
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1223525160-9887-14-git-send-email-tytso@mit.edu \
--to=tytso@mit.edu \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=cmm@us.ibm.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®