From: Chandan Rajendra <chandan@linux.vnet.ibm.com>
To: clm@fb.com, jbacik@fb.com, bo.li.liu@oracle.com, dsterba@suse.cz
Cc: Chandan Rajendra <chandan@linux.vnet.ibm.com>,
aneesh.kumar@linux.vnet.ibm.com, linux-btrfs@vger.kernel.org,
chandan@mykolab.com
Subject: [PATCH V14 14/15] Btrfs: subpagesize-blocksize: extent_clear_unlock_delalloc: Prevent page from being unlocked more than once
Date: Thu, 4 Feb 2016 12:16:21 +0530 [thread overview]
Message-ID: <1454568382-2020-15-git-send-email-chandan@linux.vnet.ibm.com> (raw)
In-Reply-To: <1454568382-2020-1-git-send-email-chandan@linux.vnet.ibm.com>
extent_clear_unlock_delalloc() can unlock a page more than once as shown
below (assume 4k as the block size and 64k as the page size).
cow_file_range
create 4k ordered extent corresponding to page offsets 0 - 4095
extent_clear_unlock_delalloc corresponding to page offsets 0 - 4095
unlock page
create 4k ordered extent corresponding to page offsets 4096 - 8191
extent_clear_unlock_delalloc corresponding to page offsets 4096 - 8191
unlock page
To prevent such a scenario this commit passes "delalloc end" to
extent_clear_unlock_delalloc() to help decide whether the page can be unlocked
or not.
NOTE: Since extent_clear_unlock_delalloc() is used by compression code as
well, the commit passes ordered extent "end" as the value for the argument
corresponding to "delalloc end" for invocations made from compression code
path. This will be fixed by a future commit that gets compression to work in
subpagesize-blocksize scenario.
Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
fs/btrfs/extent_io.c | 15 ++++++----
fs/btrfs/extent_io.h | 5 ++--
fs/btrfs/inode.c | 84 ++++++++++++++++++++++++++++++----------------------
3 files changed, 60 insertions(+), 44 deletions(-)
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 6a6fdd6..a7d4772 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -1825,9 +1825,8 @@ out_failed:
}
void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
- struct page *locked_page,
- unsigned clear_bits,
- unsigned long page_ops)
+ u64 delalloc_end, struct page *locked_page,
+ unsigned clear_bits, unsigned long page_ops)
{
struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
int ret;
@@ -1835,6 +1834,7 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
unsigned long index = start >> PAGE_CACHE_SHIFT;
unsigned long end_index = end >> PAGE_CACHE_SHIFT;
unsigned long nr_pages = end_index - index + 1;
+ u64 page_end;
int i;
clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
@@ -1871,8 +1871,13 @@ void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
if ((page_ops & PAGE_END_WRITEBACK)
&& !PagePrivate2(pages[i]))
end_page_writeback(pages[i]);
- if (page_ops & PAGE_UNLOCK)
- unlock_page(pages[i]);
+ if (page_ops & PAGE_UNLOCK) {
+ page_end = page_offset(pages[i]) +
+ PAGE_CACHE_SIZE - 1;
+ if ((page_end <= end)
+ || (end == delalloc_end))
+ unlock_page(pages[i]);
+ }
page_cache_release(pages[i]);
}
nr_pages -= ret;
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 9b777b0..bcb00f7 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -462,9 +462,8 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
- struct page *locked_page,
- unsigned bits_to_clear,
- unsigned long page_ops);
+ u64 delalloc_end, struct page *locked_page,
+ unsigned bits_to_clear, unsigned long page_ops);
struct bio *
btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
gfp_t gfp_flags);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index c6fa733..35f0e18 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -104,9 +104,10 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr);
static int btrfs_truncate(struct inode *inode);
static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent);
static noinline int cow_file_range(struct inode *inode,
- struct page *locked_page,
- u64 start, u64 end, int *page_started,
- unsigned long *nr_written, int unlock);
+ struct page *locked_page,
+ u64 start, u64 end, u64 delalloc_end,
+ int *page_started, unsigned long *nr_written,
+ int unlock);
static struct extent_map *create_pinned_em(struct inode *inode, u64 start,
u64 len, u64 orig_start,
u64 block_start, u64 block_len,
@@ -561,12 +562,13 @@ cont:
* we don't need to create any more async work items.
* Unlock and free up our temp pages.
*/
- extent_clear_unlock_delalloc(inode, start, end, NULL,
- clear_flags, PAGE_UNLOCK |
- PAGE_CLEAR_DIRTY |
- PAGE_SET_WRITEBACK |
- page_error_op |
- PAGE_END_WRITEBACK);
+ extent_clear_unlock_delalloc(inode, start, end, end,
+ NULL, clear_flags,
+ PAGE_UNLOCK
+ | PAGE_CLEAR_DIRTY
+ | PAGE_SET_WRITEBACK
+ | page_error_op
+ | PAGE_END_WRITEBACK);
goto free_pages_out;
}
}
@@ -715,6 +717,8 @@ retry:
async_extent->start,
async_extent->start +
async_extent->ram_size - 1,
+ async_extent->start +
+ async_extent->ram_size - 1,
&page_started, &nr_written, 0);
/* JDM XXX */
@@ -834,6 +838,8 @@ retry:
extent_clear_unlock_delalloc(inode, async_extent->start,
async_extent->start +
async_extent->ram_size - 1,
+ async_extent->start +
+ async_extent->ram_size - 1,
NULL, EXTENT_LOCKED | EXTENT_DELALLOC,
PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
PAGE_SET_WRITEBACK);
@@ -853,9 +859,10 @@ retry:
tree->ops->writepage_end_io_hook(p, start, end,
NULL, 0);
p->mapping = NULL;
- extent_clear_unlock_delalloc(inode, start, end, NULL, 0,
- PAGE_END_WRITEBACK |
- PAGE_SET_ERROR);
+ extent_clear_unlock_delalloc(inode, start, end, end,
+ NULL, 0,
+ PAGE_END_WRITEBACK |
+ PAGE_SET_ERROR);
free_async_extent_pages(async_extent);
}
alloc_hint = ins.objectid + ins.offset;
@@ -869,6 +876,8 @@ out_free:
extent_clear_unlock_delalloc(inode, async_extent->start,
async_extent->start +
async_extent->ram_size - 1,
+ async_extent->start +
+ async_extent->ram_size - 1,
NULL, EXTENT_LOCKED | EXTENT_DELALLOC |
EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING,
PAGE_UNLOCK | PAGE_CLEAR_DIRTY |
@@ -925,10 +934,10 @@ static u64 get_extent_allocation_hint(struct inode *inode, u64 start,
* IO when we return.
*/
static noinline int cow_file_range(struct inode *inode,
- struct page *locked_page,
- u64 start, u64 end, int *page_started,
- unsigned long *nr_written,
- int unlock)
+ struct page *locked_page,
+ u64 start, u64 end, u64 delalloc_end,
+ int *page_started, unsigned long *nr_written,
+ int unlock)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
u64 alloc_hint = 0;
@@ -964,7 +973,8 @@ static noinline int cow_file_range(struct inode *inode,
ret = cow_file_range_inline(root, inode, start, end, 0, 0,
NULL);
if (ret == 0) {
- extent_clear_unlock_delalloc(inode, start, end, NULL,
+ extent_clear_unlock_delalloc(inode, start, end,
+ delalloc_end, NULL,
EXTENT_LOCKED | EXTENT_DELALLOC |
EXTENT_DEFRAG, PAGE_UNLOCK |
PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK |
@@ -1054,9 +1064,9 @@ static noinline int cow_file_range(struct inode *inode,
page_ops = unlock ? PAGE_UNLOCK : 0;
page_ops |= PAGE_SET_PRIVATE2;
extent_ops = EXTENT_LOCKED | EXTENT_DELALLOC;
- extent_clear_unlock_delalloc(inode, start,
- start + ram_size - 1, locked_page,
- extent_ops, page_ops);
+ extent_clear_unlock_delalloc(inode, start, start + ram_size - 1,
+ delalloc_end, locked_page, extent_ops,
+ page_ops);
disk_num_bytes -= cur_alloc_size;
num_bytes -= cur_alloc_size;
alloc_hint = ins.objectid + ins.offset;
@@ -1087,8 +1097,8 @@ out_unlock:
extent_ops = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING
| EXTENT_DEFRAG;
- extent_clear_unlock_delalloc(inode, start, end, locked_page,
- extent_ops, page_ops);
+ extent_clear_unlock_delalloc(inode, start, end, delalloc_end,
+ locked_page, extent_ops, page_ops);
goto out;
}
@@ -1235,9 +1245,9 @@ static noinline int csum_exist_in_range(struct btrfs_root *root,
* blocks on disk
*/
static noinline int run_delalloc_nocow(struct inode *inode,
- struct page *locked_page,
- u64 start, u64 end, int *page_started, int force,
- unsigned long *nr_written)
+ struct page *locked_page,
+ u64 start, u64 end, int *page_started,
+ int force, unsigned long *nr_written)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_trans_handle *trans;
@@ -1263,7 +1273,8 @@ static noinline int run_delalloc_nocow(struct inode *inode,
path = btrfs_alloc_path();
if (!path) {
- extent_clear_unlock_delalloc(inode, start, end, locked_page,
+ extent_clear_unlock_delalloc(inode, start, end, end,
+ locked_page,
EXTENT_LOCKED | EXTENT_DELALLOC |
EXTENT_DO_ACCOUNTING |
EXTENT_DEFRAG, PAGE_UNLOCK |
@@ -1281,7 +1292,8 @@ static noinline int run_delalloc_nocow(struct inode *inode,
trans = btrfs_join_transaction(root);
if (IS_ERR(trans)) {
- extent_clear_unlock_delalloc(inode, start, end, locked_page,
+ extent_clear_unlock_delalloc(inode, start, end, end,
+ locked_page,
EXTENT_LOCKED | EXTENT_DELALLOC |
EXTENT_DO_ACCOUNTING |
EXTENT_DEFRAG, PAGE_UNLOCK |
@@ -1422,8 +1434,8 @@ out_check:
btrfs_release_path(path);
if (cow_start != (u64)-1) {
ret = cow_file_range(inode, locked_page,
- cow_start, found_key.offset - 1,
- page_started, nr_written, 1);
+ cow_start, found_key.offset - 1, end,
+ page_started, nr_written, 1);
if (ret) {
if (!nolock && nocow)
btrfs_end_write_no_snapshoting(root);
@@ -1483,10 +1495,10 @@ out_check:
}
extent_clear_unlock_delalloc(inode, cur_offset,
- cur_offset + num_bytes - 1,
- locked_page, EXTENT_LOCKED |
- EXTENT_DELALLOC, PAGE_UNLOCK |
- PAGE_SET_PRIVATE2);
+ cur_offset + num_bytes - 1, end,
+ locked_page, EXTENT_LOCKED |
+ EXTENT_DELALLOC, PAGE_UNLOCK |
+ PAGE_SET_PRIVATE2);
if (!nolock && nocow)
btrfs_end_write_no_snapshoting(root);
cur_offset = extent_end;
@@ -1501,7 +1513,7 @@ out_check:
}
if (cow_start != (u64)-1) {
- ret = cow_file_range(inode, locked_page, cow_start, end,
+ ret = cow_file_range(inode, locked_page, cow_start, end, end,
page_started, nr_written, 1);
if (ret)
goto error;
@@ -1513,7 +1525,7 @@ error:
ret = err;
if (ret && cur_offset < end)
- extent_clear_unlock_delalloc(inode, cur_offset, end,
+ extent_clear_unlock_delalloc(inode, cur_offset, end, end,
locked_page, EXTENT_LOCKED |
EXTENT_DELALLOC | EXTENT_DEFRAG |
EXTENT_DO_ACCOUNTING, PAGE_UNLOCK |
@@ -1561,7 +1573,7 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page,
ret = run_delalloc_nocow(inode, locked_page, start, end,
page_started, 0, nr_written);
} else if (!inode_need_compress(inode)) {
- ret = cow_file_range(inode, locked_page, start, end,
+ ret = cow_file_range(inode, locked_page, start, end, end,
page_started, nr_written, 1);
} else {
set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
--
2.1.0
next prev parent reply other threads:[~2016-02-04 6:48 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-04 6:46 [PATCH V14 00/15] Btrfs: Subpagesize-blocksize: Allow I/O on blocks whose size is less than page size Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 01/15] Btrfs: subpagesize-blocksize: Fix whole page read Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 02/15] Btrfs: subpagesize-blocksize: Fix whole page write Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 03/15] Btrfs: subpagesize-blocksize: Make sure delalloc range intersects with the locked page's range Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 04/15] Btrfs: subpagesize-blocksize: Define extent_buffer_head Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 05/15] Btrfs: subpagesize-blocksize: Read tree blocks whose size is < PAGE_SIZE Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 06/15] Btrfs: subpagesize-blocksize: Write only dirty extent buffers belonging to a page Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 07/15] Btrfs: subpagesize-blocksize: Allow mounting filesystems where sectorsize != PAGE_SIZE Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 08/15] Btrfs: subpagesize-blocksize: Deal with partial ordered extent allocations Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 09/15] Btrfs: subpagesize-blocksize: Explicitly track I/O status of blocks of an ordered extent Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 10/15] Btrfs: subpagesize-blocksize: btrfs_punch_hole: Fix uptodate blocks check Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 11/15] Btrfs: subpagesize-blocksize: Prevent writes to an extent buffer when PG_writeback flag is set Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 12/15] Revert "btrfs: fix lockups from btrfs_clear_path_blocking" Chandan Rajendra
2016-02-04 6:46 ` [PATCH V14 13/15] Btrfs: subpagesize-blocksize: Fix file defragmentation code Chandan Rajendra
2016-02-04 6:46 ` Chandan Rajendra [this message]
2016-02-04 6:46 ` [PATCH V14 15/15] Btrfs: subpagesize-blocksize: Enable dedup ioctl Chandan Rajendra
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=1454568382-2020-15-git-send-email-chandan@linux.vnet.ibm.com \
--to=chandan@linux.vnet.ibm.com \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=bo.li.liu@oracle.com \
--cc=chandan@mykolab.com \
--cc=clm@fb.com \
--cc=dsterba@suse.cz \
--cc=jbacik@fb.com \
--cc=linux-btrfs@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.