Aneesh Kumar K.V
2010-Jan-22 05:32 UTC
[PATCH -v2] btrfs: Use correct values when updating inode i_size on fallocate
Even though we allocate more, we should be updating inode i_size
as per the arguments passed
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
Changes from V1:
We should update i_size only if actual_len and cur_offset are both
larger than i_size. Otherwise if actual_len is < i_size and cur_offset >
i_size
we may end up setting wrong i_size value
fs/btrfs/inode.c | 17 ++++++++++++-----
1 files changed, 12 insertions(+), 5 deletions(-)
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5440bab..ac4cb6b 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5789,7 +5789,7 @@ out_fail:
}
static int prealloc_file_range(struct inode *inode, u64 start, u64 end,
- u64 alloc_hint, int mode)
+ u64 alloc_hint, int mode, loff_t actual_len)
{
struct btrfs_trans_handle *trans;
struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -5798,6 +5798,7 @@ static int prealloc_file_range(struct inode *inode, u64
start, u64 end,
u64 cur_offset = start;
u64 num_bytes = end - start;
int ret = 0;
+ u64 i_size;
while (num_bytes > 0) {
alloc_size = min(num_bytes, root->fs_info->max_extent);
@@ -5835,9 +5836,15 @@ static int prealloc_file_range(struct inode *inode, u64
start, u64 end,
inode->i_ctime = CURRENT_TIME;
BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
if (!(mode & FALLOC_FL_KEEP_SIZE) &&
- cur_offset > inode->i_size) {
- i_size_write(inode, cur_offset);
- btrfs_ordered_update_i_size(inode, cur_offset, NULL);
+ (actual_len > inode->i_size) &&
+ (cur_offset > inode->i_size)) {
+
+ if (cur_offset > actual_len)
+ i_size = actual_len;
+ else
+ i_size = cur_offset;
+ i_size_write(inode, i_size);
+ btrfs_ordered_update_i_size(inode, i_size, NULL);
}
ret = btrfs_update_inode(trans, root, inode);
@@ -5930,7 +5937,7 @@ static long btrfs_fallocate(struct inode *inode, int mode,
!test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
ret = prealloc_file_range(inode,
cur_offset, last_byte,
- alloc_hint, mode);
+ alloc_hint, mode, offset+len);
if (ret < 0) {
free_extent_map(em);
break;
--
1.6.6.1.383.g5a9f
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Aneesh Kumar K. V
2010-Jan-30 08:16 UTC
Re: [PATCH -v2] btrfs: Use correct values when updating inode i_size on fallocate
Hi Chris, I see that you merged an older version of this patch. But i guess it still have some issues. -anesh On Fri, 22 Jan 2010 11:02:45 +0530, "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> wrote:> Even though we allocate more, we should be updating inode i_size > as per the arguments passed > > Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> > --- > Changes from V1: > We should update i_size only if actual_len and cur_offset are both > larger than i_size. Otherwise if actual_len is < i_size and cur_offset > i_size > we may end up setting wrong i_size value > > fs/btrfs/inode.c | 17 ++++++++++++----- > 1 files changed, 12 insertions(+), 5 deletions(-) > > diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c > index 5440bab..ac4cb6b 100644 > --- a/fs/btrfs/inode.c > +++ b/fs/btrfs/inode.c > @@ -5789,7 +5789,7 @@ out_fail: > } > > static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > - u64 alloc_hint, int mode) > + u64 alloc_hint, int mode, loff_t actual_len) > { > struct btrfs_trans_handle *trans; > struct btrfs_root *root = BTRFS_I(inode)->root; > @@ -5798,6 +5798,7 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > u64 cur_offset = start; > u64 num_bytes = end - start; > int ret = 0; > + u64 i_size; > > while (num_bytes > 0) { > alloc_size = min(num_bytes, root->fs_info->max_extent); > @@ -5835,9 +5836,15 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > inode->i_ctime = CURRENT_TIME; > BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC; > if (!(mode & FALLOC_FL_KEEP_SIZE) && > - cur_offset > inode->i_size) { > - i_size_write(inode, cur_offset); > - btrfs_ordered_update_i_size(inode, cur_offset, NULL); > + (actual_len > inode->i_size) && > + (cur_offset > inode->i_size)) { > + > + if (cur_offset > actual_len) > + i_size = actual_len; > + else > + i_size = cur_offset; > + i_size_write(inode, i_size); > + btrfs_ordered_update_i_size(inode, i_size, NULL); > } > > ret = btrfs_update_inode(trans, root, inode); > @@ -5930,7 +5937,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, > !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { > ret = prealloc_file_range(inode, > cur_offset, last_byte, > - alloc_hint, mode); > + alloc_hint, mode, offset+len); > if (ret < 0) { > free_extent_map(em); > break; > -- > 1.6.6.1.383.g5a9f > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html-- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Chris Mason
2010-Feb-01 14:31 UTC
Re: [PATCH -v2] btrfs: Use correct values when updating inode i_size on fallocate
On Sat, Jan 30, 2010 at 01:46:04PM +0530, Aneesh Kumar K. V wrote:> > Hi Chris, > > I see that you merged an older version of this patch. But i guess it > still have some issues.Thanks, I missed v2 in the patchwork list. I''ll fix it up. -chris> > -anesh > On Fri, 22 Jan 2010 11:02:45 +0530, "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> wrote: > > Even though we allocate more, we should be updating inode i_size > > as per the arguments passed > > > > Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> > > --- > > Changes from V1: > > We should update i_size only if actual_len and cur_offset are both > > larger than i_size. Otherwise if actual_len is < i_size and cur_offset > i_size > > we may end up setting wrong i_size value > > > > fs/btrfs/inode.c | 17 ++++++++++++----- > > 1 files changed, 12 insertions(+), 5 deletions(-) > > > > diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c > > index 5440bab..ac4cb6b 100644 > > --- a/fs/btrfs/inode.c > > +++ b/fs/btrfs/inode.c > > @@ -5789,7 +5789,7 @@ out_fail: > > } > > > > static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > > - u64 alloc_hint, int mode) > > + u64 alloc_hint, int mode, loff_t actual_len) > > { > > struct btrfs_trans_handle *trans; > > struct btrfs_root *root = BTRFS_I(inode)->root; > > @@ -5798,6 +5798,7 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > > u64 cur_offset = start; > > u64 num_bytes = end - start; > > int ret = 0; > > + u64 i_size; > > > > while (num_bytes > 0) { > > alloc_size = min(num_bytes, root->fs_info->max_extent); > > @@ -5835,9 +5836,15 @@ static int prealloc_file_range(struct inode *inode, u64 start, u64 end, > > inode->i_ctime = CURRENT_TIME; > > BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC; > > if (!(mode & FALLOC_FL_KEEP_SIZE) && > > - cur_offset > inode->i_size) { > > - i_size_write(inode, cur_offset); > > - btrfs_ordered_update_i_size(inode, cur_offset, NULL); > > + (actual_len > inode->i_size) && > > + (cur_offset > inode->i_size)) { > > + > > + if (cur_offset > actual_len) > > + i_size = actual_len; > > + else > > + i_size = cur_offset; > > + i_size_write(inode, i_size); > > + btrfs_ordered_update_i_size(inode, i_size, NULL); > > } > > > > ret = btrfs_update_inode(trans, root, inode); > > @@ -5930,7 +5937,7 @@ static long btrfs_fallocate(struct inode *inode, int mode, > > !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { > > ret = prealloc_file_range(inode, > > cur_offset, last_byte, > > - alloc_hint, mode); > > + alloc_hint, mode, offset+len); > > if (ret < 0) { > > free_extent_map(em); > > break; > > -- > > 1.6.6.1.383.g5a9f > > > > -- > > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html-- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Reasonably Related Threads
- [PATCH V2 11/12] Btrfs: Pre-allocate space for data relocation
- [PATCH] Btrfs: fix typo in fallocate to make it honor actual size
- [PATCH 7/8] Make fallocate(2) more ENOSPC friendly
- [PATCH 03/12] Btrfs: Rewrite btrfs_drop_extents
- [PATCH] Btrfs: fix broken nocow after balance