summaryrefslogtreecommitdiff
path: root/fs/ocfs2/aops.c
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2015-09-05 14:27:38 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2015-09-05 14:27:38 -0700
commit6c0f568e84a3cfc775682311d65205462c3f3bc1 (patch)
tree5105a137a9ea2459d55e895d3c096bbd31274724 /fs/ocfs2/aops.c
parentc82199061009d1561e31e17fca5e47a87cb7ff4c (diff)
parent559ec2f8fd50981821621f52db5e1a8ffcf8d792 (diff)
Merge branch 'akpm' (patches from Andrew)
Merge patch-bomb from Andrew Morton: - a few misc things - Andy's "ambient capabilities" - fs/nofity updates - the ocfs2 queue - kernel/watchdog.c updates and feature work. - some of MM. Includes Andrea's userfaultfd feature. [ Hadn't noticed that userfaultfd was 'default y' when applying the patches, so that got fixed in this merge instead. We do _not_ mark new features that nobody uses yet 'default y' - Linus ] * emailed patches from Andrew Morton <akpm@linux-foundation.org>: (118 commits) mm/hugetlb.c: make vma_has_reserves() return bool mm/madvise.c: make madvise_behaviour_valid() return bool mm/memory.c: make tlb_next_batch() return bool mm/dmapool.c: change is_page_busy() return from int to bool mm: remove struct node_active_region mremap: simplify the "overlap" check in mremap_to() mremap: don't do uneccesary checks if new_len == old_len mremap: don't do mm_populate(new_addr) on failure mm: move ->mremap() from file_operations to vm_operations_struct mremap: don't leak new_vma if f_op->mremap() fails mm/hugetlb.c: make vma_shareable() return bool mm: make GUP handle pfn mapping unless FOLL_GET is requested mm: fix status code which move_pages() returns for zero page mm: memcontrol: bring back the VM_BUG_ON() in mem_cgroup_swapout() genalloc: add support of multiple gen_pools per device genalloc: add name arg to gen_pool_get() and devm_gen_pool_create() mm/memblock: WARN_ON when nid differs from overlap region Documentation/features/vm: add feature description and arch support status for batched TLB flush after unmap mm: defer flush of writable TLB entries mm: send one IPI per CPU to TLB flush all entries after unmapping pages ...
Diffstat (limited to 'fs/ocfs2/aops.c')
-rw-r--r--fs/ocfs2/aops.c54
1 files changed, 41 insertions, 13 deletions
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 0f5fd9db8194..64b11d90eca6 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -227,7 +227,7 @@ int ocfs2_read_inline_data(struct inode *inode, struct page *page,
struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data;
if (!(le16_to_cpu(di->i_dyn_features) & OCFS2_INLINE_DATA_FL)) {
- ocfs2_error(inode->i_sb, "Inode %llu lost inline data flag",
+ ocfs2_error(inode->i_sb, "Inode %llu lost inline data flag\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno);
return -EROFS;
}
@@ -237,7 +237,7 @@ int ocfs2_read_inline_data(struct inode *inode, struct page *page,
if (size > PAGE_CACHE_SIZE ||
size > ocfs2_max_inline_data_with_xattr(inode->i_sb, di)) {
ocfs2_error(inode->i_sb,
- "Inode %llu has with inline data has bad size: %Lu",
+ "Inode %llu has with inline data has bad size: %Lu\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno,
(unsigned long long)size);
return -EROFS;
@@ -533,10 +533,14 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
inode_blocks = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
+ down_read(&OCFS2_I(inode)->ip_alloc_sem);
+
/* This figures out the size of the next contiguous block, and
* our logical offset */
ret = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno,
&contig_blocks, &ext_flags);
+ up_read(&OCFS2_I(inode)->ip_alloc_sem);
+
if (ret) {
mlog(ML_ERROR, "get_blocks() failed iblock=%llu\n",
(unsigned long long)iblock);
@@ -557,6 +561,8 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
alloc_locked = 1;
+ down_write(&OCFS2_I(inode)->ip_alloc_sem);
+
/* fill hole, allocate blocks can't be larger than the size
* of the hole */
clusters_to_alloc = ocfs2_clusters_for_bytes(inode->i_sb, len);
@@ -569,6 +575,7 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
ret = ocfs2_extend_allocation(inode, cpos,
clusters_to_alloc, 0);
if (ret < 0) {
+ up_write(&OCFS2_I(inode)->ip_alloc_sem);
mlog_errno(ret);
goto bail;
}
@@ -576,11 +583,13 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
ret = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno,
&contig_blocks, &ext_flags);
if (ret < 0) {
+ up_write(&OCFS2_I(inode)->ip_alloc_sem);
mlog(ML_ERROR, "get_blocks() failed iblock=%llu\n",
(unsigned long long)iblock);
ret = -EIO;
goto bail;
}
+ up_write(&OCFS2_I(inode)->ip_alloc_sem);
}
/*
@@ -627,10 +636,13 @@ static void ocfs2_dio_end_io(struct kiocb *iocb,
mutex_unlock(&OCFS2_I(inode)->ip_unaligned_aio);
}
- ocfs2_iocb_clear_rw_locked(iocb);
+ /* Let rw unlock to be done later to protect append direct io write */
+ if (offset + bytes <= i_size_read(inode)) {
+ ocfs2_iocb_clear_rw_locked(iocb);
- level = ocfs2_iocb_rw_locked_level(iocb);
- ocfs2_rw_unlock(inode, level);
+ level = ocfs2_iocb_rw_locked_level(iocb);
+ ocfs2_rw_unlock(inode, level);
+ }
}
static int ocfs2_releasepage(struct page *page, gfp_t wait)
@@ -832,12 +844,17 @@ static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb,
/* zeroing out the previously allocated cluster tail
* that but not zeroed */
- if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)))
+ if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) {
+ down_read(&OCFS2_I(inode)->ip_alloc_sem);
ret = ocfs2_direct_IO_zero_extend(osb, inode, offset,
zero_len_tail, cluster_align_tail);
- else
+ up_read(&OCFS2_I(inode)->ip_alloc_sem);
+ } else {
+ down_write(&OCFS2_I(inode)->ip_alloc_sem);
ret = ocfs2_direct_IO_extend_no_holes(osb, inode,
offset);
+ up_write(&OCFS2_I(inode)->ip_alloc_sem);
+ }
if (ret < 0) {
mlog_errno(ret);
ocfs2_inode_unlock(inode, 1);
@@ -857,7 +874,8 @@ static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb,
written = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter,
offset, ocfs2_direct_IO_get_blocks,
ocfs2_dio_end_io, NULL, 0);
- if (unlikely(written < 0)) {
+ /* overwrite aio may return -EIOCBQUEUED, and it is not an error */
+ if ((written < 0) && (written != -EIOCBQUEUED)) {
loff_t i_size = i_size_read(inode);
if (offset + count > i_size) {
@@ -876,12 +894,14 @@ static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb,
ocfs2_inode_unlock(inode, 1);
brelse(di_bh);
+ di_bh = NULL;
goto clean_orphan;
}
}
ocfs2_inode_unlock(inode, 1);
brelse(di_bh);
+ di_bh = NULL;
ret = jbd2_journal_force_commit(journal);
if (ret < 0)
@@ -936,10 +956,12 @@ clean_orphan:
if (tmp_ret < 0) {
ret = tmp_ret;
mlog_errno(ret);
+ brelse(di_bh);
goto out;
}
ocfs2_inode_unlock(inode, 1);
+ brelse(di_bh);
tmp_ret = jbd2_journal_force_commit(journal);
if (tmp_ret < 0) {
@@ -2185,10 +2207,7 @@ try_again:
if (ret)
goto out_commit;
}
- /*
- * We don't want this to fail in ocfs2_write_end(), so do it
- * here.
- */
+
ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), wc->w_di_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (ret) {
@@ -2345,7 +2364,7 @@ int ocfs2_write_end_nolock(struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
- int i;
+ int i, ret;
unsigned from, to, start = pos & (PAGE_CACHE_SIZE - 1);
struct inode *inode = mapping->host;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
@@ -2354,6 +2373,14 @@ int ocfs2_write_end_nolock(struct address_space *mapping,
handle_t *handle = wc->w_handle;
struct page *tmppage;
+ ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), wc->w_di_bh,
+ OCFS2_JOURNAL_ACCESS_WRITE);
+ if (ret) {
+ copied = ret;
+ mlog_errno(ret);
+ goto out;
+ }
+
if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
ocfs2_write_end_inline(inode, pos, len, &copied, di, wc);
goto out_write_size;
@@ -2409,6 +2436,7 @@ out_write_size:
ocfs2_update_inode_fsync_trans(handle, inode, 1);
ocfs2_journal_dirty(handle, wc->w_di_bh);
+out:
/* unlock pages before dealloc since it needs acquiring j_trans_barrier
* lock, or it will cause a deadlock since journal commit threads holds
* this lock and will ask for the page lock when flushing the data.