From: Chandan Rajendra <chandan@linux.vnet.ibm.com>
To: linux-btrfs@vger.kernel.org
Cc: Chandan Rajendra <chandan@linux.vnet.ibm.com>,
dsterba@suse.cz, clm@fb.com, jbacik@fb.com, chandan@mykolab.com,
aneesh.kumar@linux.vnet.ibm.com
Subject: [PATCH V17 17/18] Btrfs: subpage-blocksize: Make file extent relocate code subpage blocksize aware
Date: Thu, 14 Apr 2016 19:08:26 +0530 [thread overview]
Message-ID: <1460641107-13834-18-git-send-email-chandan@linux.vnet.ibm.com> (raw)
In-Reply-To: <1460641107-13834-1-git-send-email-chandan@linux.vnet.ibm.com>
The file extent relocation code currently assumes blocksize to be same
as PAGE_SIZE. This commit adds code to support subpage blocksize
scenario.
Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
fs/btrfs/relocation.c | 73 +++++++++++++++++++++++++++++++++------------------
1 file changed, 48 insertions(+), 25 deletions(-)
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 35fd62e..61a6414 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3103,14 +3103,19 @@ static int relocate_file_extent_cluster(struct inode *inode,
{
u64 page_start;
u64 page_end;
+ u64 block_start;
+ u64 block_end;
u64 offset = BTRFS_I(inode)->index_cnt;
+ u64 blocksize = BTRFS_I(inode)->root->sectorsize;
unsigned long index;
unsigned long last_index;
struct page *page;
struct file_ra_state *ra;
gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping);
+ int nr_blocks;
int nr = 0;
int ret = 0;
+ int i;
if (!cluster->nr)
return 0;
@@ -3130,13 +3135,13 @@ static int relocate_file_extent_cluster(struct inode *inode,
if (ret)
goto out;
+ page_start = cluster->start - offset;
+ page_end = min_t(u64, page_start + PAGE_SIZE - 1,
+ cluster->end - offset);
+
index = (cluster->start - offset) >> PAGE_SHIFT;
last_index = (cluster->end - offset) >> PAGE_SHIFT;
while (index <= last_index) {
- ret = btrfs_delalloc_reserve_metadata(inode, PAGE_SIZE);
- if (ret)
- goto out;
-
page = find_lock_page(inode->i_mapping, index);
if (!page) {
page_cache_sync_readahead(inode->i_mapping,
@@ -3145,8 +3150,6 @@ static int relocate_file_extent_cluster(struct inode *inode,
page = find_or_create_page(inode->i_mapping, index,
mask);
if (!page) {
- btrfs_delalloc_release_metadata(inode,
- PAGE_SIZE);
ret = -ENOMEM;
goto out;
}
@@ -3164,42 +3167,62 @@ static int relocate_file_extent_cluster(struct inode *inode,
if (!PageUptodate(page)) {
unlock_page(page);
put_page(page);
- btrfs_delalloc_release_metadata(inode,
- PAGE_SIZE);
ret = -EIO;
goto out;
}
}
- page_start = page_offset(page);
- page_end = page_start + PAGE_SIZE - 1;
+ nr_blocks = (page_end + 1 - page_start) >> inode->i_blkbits;
+ i = 0;
- lock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end);
+ while (i < nr_blocks) {
+ ret = btrfs_delalloc_reserve_metadata(inode, blocksize);
+ if (ret) {
+ if (i)
+ set_page_dirty(page);
+ unlock_page(page);
+ put_page(page);
+ goto out;
+ }
+
+ block_start = page_start + (i << inode->i_blkbits);
+ block_end = block_start + blocksize - 1;
+
+ lock_extent(&BTRFS_I(inode)->io_tree, block_start,
+ block_end);
+
+ set_page_extent_mapped(page);
- set_page_extent_mapped(page);
+ if (nr < cluster->nr &&
+ block_start + offset == cluster->boundary[nr]) {
+ set_extent_bits(&BTRFS_I(inode)->io_tree,
+ block_start, block_end,
+ EXTENT_BOUNDARY, GFP_NOFS);
+ nr++;
+ }
+
+ btrfs_set_extent_delalloc(inode, block_start,
+ block_end, NULL);
+ set_page_blks_state(page,
+ 1 << BLK_STATE_DIRTY | 1 << BLK_STATE_UPTODATE,
+ block_start, block_end);
- if (nr < cluster->nr &&
- page_start + offset == cluster->boundary[nr]) {
- set_extent_bits(&BTRFS_I(inode)->io_tree,
- page_start, page_end,
- EXTENT_BOUNDARY, GFP_NOFS);
- nr++;
+ unlock_extent(&BTRFS_I(inode)->io_tree,
+ block_start, block_end);
+ i++;
}
- btrfs_set_extent_delalloc(inode, page_start, page_end, NULL);
- set_page_blks_state(page,
- 1 << BLK_STATE_DIRTY | 1 << BLK_STATE_UPTODATE,
- page_start, page_end);
set_page_dirty(page);
-
- unlock_extent(&BTRFS_I(inode)->io_tree,
- page_start, page_end);
unlock_page(page);
put_page(page);
index++;
balance_dirty_pages_ratelimited(inode->i_mapping);
btrfs_throttle(BTRFS_I(inode)->root);
+
+ page_start += PAGE_SIZE;
+ page_end = min_t(u64, page_start + PAGE_SIZE - 1,
+ cluster->end - offset);
}
WARN_ON(nr != cluster->nr);
out:
--
2.1.0
next prev parent reply other threads:[~2016-04-14 13:39 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-14 13:38 [PATCH V17 00/18] Allow I/O on blocks whose size is less than page size Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 01/18] Btrfs: subpage-blocksize: Fix whole page read Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 02/18] Btrfs: subpage-blocksize: Fix whole page write Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 03/18] Btrfs: subpage-blocksize: Make sure delalloc range intersects with the locked page's range Chandan Rajendra
2016-04-17 12:24 ` [PATCH] " Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 04/18] Btrfs: subpage-blocksize: Define extent_buffer_head Chandan Rajendra
2016-04-14 15:49 ` kbuild test robot
2016-04-17 12:24 ` [PATCH] " Chandan Rajendra
2016-04-17 12:53 ` kbuild test robot
2016-04-18 10:00 ` kbuild test robot
2016-04-14 13:38 ` [PATCH V17 05/18] Btrfs: subpage-blocksize: Read tree blocks whose size is < PAGE_CACHE_SIZE Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 06/18] Btrfs: subpage-blocksize: Write only dirty extent buffers belonging to a page Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 07/18] Btrfs: subpage-blocksize: Allow mounting filesystems where sectorsize < PAGE_SIZE Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 08/18] Btrfs: subpage-blocksize: Deal with partial ordered extent allocations Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 09/18] Btrfs: subpage-blocksize: Explicitly track I/O status of blocks of an ordered extent Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 10/18] Btrfs: subpage-blocksize: btrfs_punch_hole: Fix uptodate blocks check Chandan Rajendra
2016-04-14 16:04 ` kbuild test robot
2016-04-17 12:15 ` [PATCH] " Chandan Rajendra
2016-04-17 12:24 ` Chandan Rajendra
2016-04-17 12:38 ` kbuild test robot
2016-04-18 8:35 ` kbuild test robot
2016-04-14 13:38 ` [PATCH V17 11/18] Btrfs: subpage-blocksize: Prevent writes to an extent buffer when PG_writeback flag is set Chandan Rajendra
2016-04-17 12:25 ` [PATCH] " Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 12/18] Revert "btrfs: fix lockups from btrfs_clear_path_blocking" Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 13/18] Btrfs: subpage-blocksize: Fix file defragmentation code Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 14/18] Btrfs: subpage-blocksize: extent_clear_unlock_delalloc: Prevent page from being unlocked more than once Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 15/18] Btrfs: subpage-blocksize: Enable dedupe ioctl Chandan Rajendra
2016-04-14 13:38 ` [PATCH V17 16/18] Btrfs: btrfs_clone: Flush dirty blocks of a page that do not map the clone range Chandan Rajendra
2016-04-14 13:38 ` Chandan Rajendra [this message]
2016-04-14 13:38 ` [PATCH V17 18/18] Btrfs: subpage-blocksize: __btrfs_lookup_bio_sums: Set offset when moving to a new bio_vec Chandan Rajendra
2016-04-17 12:32 ` [PATCH V17 00/18] Allow I/O on blocks whose size is less than page size 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=1460641107-13834-18-git-send-email-chandan@linux.vnet.ibm.com \
--to=chandan@linux.vnet.ibm.com \
--cc=aneesh.kumar@linux.vnet.ibm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).