Now fallocate will do accurate qgroup reserve space check, unlike old
method, which will always reserve the whole length of the range.
With this patch, fallocate will:
1) Iterate the desired range and mark in data rsv map
Only range which is going to be allocated will be recorded in data
rsv
For NOCOW and inline case, there will be no delayed_ref created for
them, so we should free their reserved data space at proper
time(finish_ordered_io for NOCOW and cow_file_inline for inline).
Signed-off-by: Qu Wenruo
---
fs/btrfs/extent-tree.c | 7 ++-
fs/btrfs/inode.c | 15
Introduce the new function to search the most left reserve range in a
reserve map.
It provides the basis for later reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 36
1 file changed, 36 insertions(+)
diff --git a/fs/btrfs/qgroup.c b/
Introduce functions btrfs_qgroup_release/free_data() to release/free
reserved data range.
Release means, just remove the data range from data rsv map, but doesn't
free the reserved space.
This is for normal buffered write case, when data is written into disc
and its metadata is added into tree, it
Add new structures and functions for new qgroup reserve implement dirty
phase.
Which will focus on avoiding over-reserve as in that case, which means
for already reserved dirty space range, we won't reserve space again.
This patch adds the needed structure declaration and comments.
Signed-off-by:
This new function will do all the hard work to reserve precious space
for a write.
The overall work flow will be the following.
File A already has some dirty pages:
0 4K 8K 12K 16K
|///| |///|
And then, someone want to write some data into range [4K, 16K).
Add new function btrfs_add_delayed_qgroup_reserve() function to record
how much space is reserved for that extent.
As btrfs only accounts qgroup at run_delayed_refs() time, so newly
allocated extent should keep the reserved space until then.
So add needed function with related members to do it.
New functions btrfs_qgroup_init/free_data_rsv_map() to init/free data
reserve map.
Data reserve map is used to mark which range already holds reserved
space, to avoid current reserved space leak.
Signed-off-by: Qu Wenruo
---
fs/btrfs/btrfs_inode.h | 2 ++
fs/btrfs/inode.c | 10 +++
f
Cleanup the old facilities which use old btrfs_qgroup_reserve() function
call, replace them with the newer version, and remove the "__" prefix in
them.
Also, make btrfs_qgroup_reserve/free() functions private, as they are
now only used inside qgroup codes.
Now, the whole btrfs qgroup is swithed t
Introduce new function reserve_data_range().
This function will find non-overlap range and to insert it into reserve
map using previously introduced functions.
This provides the basis for later per inode reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 92
New function insert_data_ranges() will insert non-overlap reserve ranges
into reserve map.
It provides the basis for later qgroup reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 124 ++
1 file changed, 124 insertions(+)
Qgroup reserved space needs to be released from inode dirty map and get
freed at different timing:
1) Release when the metadata is written into tree
After corresponding metadata is written into tree, any newer write will
be COWed(don't include NOCOW case yet).
So we must release its range from ino
Introduce new function release_data_range() to release reserved ranges.
It will iterate through all existing ranges and remove/shrink them.
Note this function will not free reserved space, as the range can be
released in the following conditions:
1) The dirty range gets written to disk.
In this
As we have the new metadata reservation functions, use them to replace
the old btrfs_qgroup_reserve() call for metadata.
Signed-off-by: Qu Wenruo
---
fs/btrfs/extent-tree.c | 14 ++
fs/btrfs/transaction.c | 34 ++
fs/btrfs/transaction.h | 1 -
3 files
Introduce new functions btrfs_qgroup_reserve/free_meta() to reserve/free
metadata reserved space.
Signed-off-by: Qu Wenruo
---
fs/btrfs/ctree.h | 3 +++
fs/btrfs/disk-io.c | 1 +
fs/btrfs/qgroup.c | 40
fs/btrfs/qgroup.h | 4
4 files changed,
Add new function __btrfs_check_data_free_space() to do precious space
reservation.
The new function will replace old btrfs_check_data_free_space(), but
until all the change is done, let's just use the new name.
Also, export internal use function btrfs_alloc_data_chunk_ondemand(), as
now qgroup re
Use new check_data_free_space for buffered write and inode cache.
For buffered write case, as nodatacow write won't increase quota account,
so unlike old behavior which does reserve before check nocow, now we
check nocow first and then only reserve data if we can't do nocow write.
Signed-off-by:
Add new version of btrfs_delalloc_reserve_space() function, which
supports accurate qgroup reserve.
Signed-off-by: Qu Wenruo
---
fs/btrfs/ctree.h | 1 +
fs/btrfs/extent-tree.c | 38 ++
2 files changed, 39 insertions(+)
diff --git a/fs/btrfs/ctree.h b/f
Use new __btrfs_delalloc_reserve_space to reserve space.
Signed-off-by: Qu Wenruo
---
fs/btrfs/inode-map.c | 2 +-
fs/btrfs/inode.c | 16 ++--
fs/btrfs/ioctl.c | 5 +++--
3 files changed, 14 insertions(+), 9 deletions(-)
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-m
[[BUG]]
One of the most common case to trigger the bug is the following method:
1) Enable quota
2) Limit excl of qgroup 5 to 16M
3) Write [0,2M) of a file inside subvol 5 10 times without sync
EQUOT will be triggered at about the 8th write.
[[CAUSE]]
The problem is caused by the fact that qgroup
Qu Wenruo wrote on 2015/09/09 09:34 +0800:
Hi David,
Sorry for the late reply, but I noticed something interesting.
David Sterba wrote on 2015/06/22 17:00 +0200:
Hi,
btrfs-progs 4.1 have been released (in time with kernel 4.1). Unusual
load of
changes.
Fixed since rc1:
- uuid rewrite pr
Hi, Qu,
On 2015/09/08 18:25, Qu Wenruo wrote:
> Cleanup the old facilities which use old btrfs_qgroup_reserve() function
> call, replace them with the newer version, and remove the "__" prefix in
> them.
>
> Also, make btrfs_qgroup_reserve/free() functions private, as they are
> now only used ins
Hi, Qu,
On 2015/09/08 18:22, Qu Wenruo wrote:
> Now fallocate will do accurate qgroup reserve space check, unlike old
> method, which will always reserve the whole length of the range.
>
> With this patch, fallocate will:
> 1) Iterate the desired range and mark in data rsv map
> Only range wh
Zhao Lei wrote on 2015/09/04 21:23 +0800:
To check is btrfs-convert create bad filesystem with
leaf accross stripes.
It is happened in progs version <=v4.1.2, and fixed by patch titled:
btrfs: convert: Avoid allocating metadata extent crossing stripe boundary
which was merged in v4.2.
Notice
Tsutomu Itoh wrote on 2015/09/09 10:21 +0900:
Hi, Qu,
On 2015/09/08 18:08, Qu Wenruo wrote:
Qgroup reserved space needs to be released from inode dirty map and get
freed at different timing:
1) Release when the metadata is written into tree
After corresponding metadata is written into tree,
On 09/09/2015 03:34 AM, Hugo Mills wrote:
On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
Hi,
Currently i have a raid1 configuration on two disks where one of them
is failing.
But since:
btrfs fi df /mnt/disk/
Data, RAID1: total=858.00GiB, used=638.16GiB
Data, single: total=1.00
Hi, Qu,
On 2015/09/08 18:22, Qu Wenruo wrote:
> Add new function __btrfs_check_data_free_space() to do precious space
> reservation.
>
> The new function will replace old btrfs_check_data_free_space(), but
> until all the change is done, let's just use the new name.
>
> Also, export internal use
Hi David,
Sorry for the late reply, but I noticed something interesting.
David Sterba wrote on 2015/06/22 17:00 +0200:
Hi,
btrfs-progs 4.1 have been released (in time with kernel 4.1). Unusual load of
changes.
Fixed since rc1:
- uuid rewrite prints the correct original UUID
- map-logica
Hi, Qu,
On 2015/09/08 18:08, Qu Wenruo wrote:
> Qgroup reserved space needs to be released from inode dirty map and get
> freed at different timing:
>
> 1) Release when the metadata is written into tree
> After corresponding metadata is written into tree, any newer write will
> be COWed(don't inc
On Tue, Sep 08, 2015 at 04:08:43PM -0700, Andy Lutomirski wrote:
> On Tue, Sep 8, 2015 at 3:39 PM, Darrick J. Wong
> wrote:
> > On Tue, Sep 08, 2015 at 02:45:39PM -0700, Andy Lutomirski wrote:
> >> On Tue, Sep 8, 2015 at 2:29 PM, Darrick J. Wong
> >> wrote:
> >> > On Tue, Sep 08, 2015 at 09:03:
Hi, Qu,
On 2015/09/08 18:08, Qu Wenruo wrote:
> Introduce functions btrfs_qgroup_release/free_data() to release/free
> reserved data range.
>
> Release means, just remove the data range from data rsv map, but doesn't
> free the reserved space.
> This is for normal buffered write case, when data i
Hi, Qu,
On 2015/09/08 18:01, Qu Wenruo wrote:
> New function insert_data_ranges() will insert non-overlap reserve ranges
> into reserve map.
>
> It provides the basis for later qgroup reserve map implement.
>
> Signed-off-by: Qu Wenruo
> ---
> fs/btrfs/qgroup.c | 124
> ++
Hi, Qu,
On 2015/09/08 17:56, Qu Wenruo wrote:
> Add new structures and functions for new qgroup reserve implement dirty
> phase.
> Which will focus on avoiding over-reserve as in that case, which means
> for already reserved dirty space range, we won't reserve space again.
>
> This patch adds the
On Tue, Sep 8, 2015 at 3:39 PM, Darrick J. Wong wrote:
> On Tue, Sep 08, 2015 at 02:45:39PM -0700, Andy Lutomirski wrote:
>> On Tue, Sep 8, 2015 at 2:29 PM, Darrick J. Wong
>> wrote:
>> > On Tue, Sep 08, 2015 at 09:03:09PM +0100, Pádraig Brady wrote:
>> >> On 08/09/15 20:10, Andy Lutomirski wrot
On Tue, Sep 08, 2015 at 02:45:39PM -0700, Andy Lutomirski wrote:
> On Tue, Sep 8, 2015 at 2:29 PM, Darrick J. Wong
> wrote:
> > On Tue, Sep 08, 2015 at 09:03:09PM +0100, Pádraig Brady wrote:
> >> On 08/09/15 20:10, Andy Lutomirski wrote:
> >> > On Tue, Sep 8, 2015 at 11:23 AM, Anna Schumaker
> >>
On Tue, Sep 8, 2015 at 2:29 PM, Darrick J. Wong wrote:
> On Tue, Sep 08, 2015 at 09:03:09PM +0100, Pádraig Brady wrote:
>> On 08/09/15 20:10, Andy Lutomirski wrote:
>> > On Tue, Sep 8, 2015 at 11:23 AM, Anna Schumaker
>> > wrote:
>> >> On 09/08/2015 11:21 AM, Pádraig Brady wrote:
>> >>> I see cop
On Tue, Sep 08, 2015 at 09:03:09PM +0100, Pádraig Brady wrote:
> On 08/09/15 20:10, Andy Lutomirski wrote:
> > On Tue, Sep 8, 2015 at 11:23 AM, Anna Schumaker
> > wrote:
> >> On 09/08/2015 11:21 AM, Pádraig Brady wrote:
> >>> I see copy_file_range() is a reflink() on BTRFS?
> >>> That's a bit surp
Rather than have three separate if() statements for the same outcome
we should just OR them together in the same if() statement.
Signed-off-by: Alexandru Moise <00moses.alexande...@gmail.com>
---
fs/btrfs/extent-tree.c | 10 +++---
1 file changed, 3 insertions(+), 7 deletions(-)
diff --git a
On 09/08/2015 04:45 PM, Darrick J. Wong wrote:
> On Tue, Sep 08, 2015 at 11:08:03AM -0400, Anna Schumaker wrote:
>> On 09/05/2015 04:33 AM, Al Viro wrote:
>>> On Fri, Sep 04, 2015 at 04:25:27PM -0600, Andreas Dilger wrote:
>>>
This is a bit of a surprising result, since in my testing in the
>>
On Tue, Sep 08, 2015 at 11:08:03AM -0400, Anna Schumaker wrote:
> On 09/05/2015 04:33 AM, Al Viro wrote:
> > On Fri, Sep 04, 2015 at 04:25:27PM -0600, Andreas Dilger wrote:
> >
> >> This is a bit of a surprising result, since in my testing in the
> >> past, copy_{to/from}_user() is a major consume
On Tue, Sep 08, 2015 at 10:33:54PM +0200, Ian Kumlien wrote:
> On 8 September 2015 at 22:28, Hugo Mills wrote:
> > On Tue, Sep 08, 2015 at 02:17:55PM -0600, Chris Murphy wrote:
> >> On Tue, Sep 8, 2015 at 2:13 PM, Ian Kumlien wrote:
> >> > On 8 September 2015 at 22:08, Chris Murphy
> >> > wrote
On Tue, Sep 08, 2015 at 11:04:03AM -0400, Anna Schumaker wrote:
> On 09/04/2015 05:38 PM, Darrick J. Wong wrote:
> > On Fri, Sep 04, 2015 at 04:17:03PM -0400, Anna Schumaker wrote:
> >> copy_file_range() is a new system call for copying ranges of data
> >> completely in the kernel. This gives file
On 8 September 2015 at 22:28, Hugo Mills wrote:
> On Tue, Sep 08, 2015 at 02:17:55PM -0600, Chris Murphy wrote:
>> On Tue, Sep 8, 2015 at 2:13 PM, Ian Kumlien wrote:
>> > On 8 September 2015 at 22:08, Chris Murphy wrote:
>> >> On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
[--8<--]
>> > -
On Tue, Sep 08, 2015 at 02:17:55PM -0600, Chris Murphy wrote:
> On Tue, Sep 8, 2015 at 2:13 PM, Ian Kumlien wrote:
> > On 8 September 2015 at 22:08, Chris Murphy wrote:
> >> On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
> >
> > [--8<--]
> >
> >>> Someone thought they were done too early, on
On 8 September 2015 at 22:17, Chris Murphy wrote:
> On Tue, Sep 8, 2015 at 2:13 PM, Ian Kumlien wrote:
>> On 8 September 2015 at 22:08, Chris Murphy wrote:
>>> On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
>>
>> [--8<--]
>>
Someone thought they were done too early, only one disk => re
On Tue, Sep 8, 2015 at 2:13 PM, Ian Kumlien wrote:
> On 8 September 2015 at 22:08, Chris Murphy wrote:
>> On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
>
> [--8<--]
>
>>> Someone thought they were done too early, only one disk => read only
>>> mount. But, readonly mount => no balance.
>>>
>
On 8 September 2015 at 22:08, Chris Murphy wrote:
> On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
[--8<--]
>> Someone thought they were done too early, only one disk => read only
>> mount. But, readonly mount => no balance.
>>
>> I think something is wrong
>>
>> btrfs balance start -dc
On Tue, Sep 8, 2015 at 2:00 PM, Ian Kumlien wrote:
> On 8 September 2015 at 21:55, Ian Kumlien wrote:
>> On 8 September 2015 at 21:43, Ian Kumlien wrote:
>>> On 8 September 2015 at 21:34, Hugo Mills wrote:
On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
>> [--8<--]
>>
On 08/09/15 20:10, Andy Lutomirski wrote:
> On Tue, Sep 8, 2015 at 11:23 AM, Anna Schumaker
> wrote:
>> On 09/08/2015 11:21 AM, Pádraig Brady wrote:
>>> I see copy_file_range() is a reflink() on BTRFS?
>>> That's a bit surprising, as it avoids the copy completely.
>>> cp(1) for example considered
> -Original Message-
> From: Austin S Hemmelgarn [mailto:ahferro...@gmail.com]
> Sent: Tuesday, September 8, 2015 7:56 AM
> Subject: Re: mkfs.btrfs cannot find rotational file for SSD detection for
> a pmem device
>
> On 2015-09-06 13:51, Elliott, Robert (Persistent Memory) wrote:
...
> >
On 8 September 2015 at 21:55, Ian Kumlien wrote:
> On 8 September 2015 at 21:43, Ian Kumlien wrote:
>> On 8 September 2015 at 21:34, Hugo Mills wrote:
>>> On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
> [--8<--]
>
>>>Physically removing it is the way to go (or disabling it usi
On 8 September 2015 at 21:43, Ian Kumlien wrote:
> On 8 September 2015 at 21:34, Hugo Mills wrote:
>> On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
[--8<--]
>>Physically removing it is the way to go (or disabling it using echo
>> offline >/sys/block/sda/device/state). Once you
On 8 September 2015 at 21:34, Hugo Mills wrote:
> On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
>> Hi,
>>
>> Currently i have a raid1 configuration on two disks where one of them
>> is failing.
>>
>> But since:
>> btrfs fi df /mnt/disk/
>> Data, RAID1: total=858.00GiB, used=638.16Gi
On Tue, Sep 08, 2015 at 09:18:05PM +0200, Ian Kumlien wrote:
> Hi,
>
> Currently i have a raid1 configuration on two disks where one of them
> is failing.
>
> But since:
> btrfs fi df /mnt/disk/
> Data, RAID1: total=858.00GiB, used=638.16GiB
> Data, single: total=1.00GiB, used=256.00KiB
> System,
Hi,
Currently i have a raid1 configuration on two disks where one of them
is failing.
But since:
btrfs fi df /mnt/disk/
Data, RAID1: total=858.00GiB, used=638.16GiB
Data, single: total=1.00GiB, used=256.00KiB
System, RAID1: total=32.00MiB, used=132.00KiB
Metadata, RAID1: total=4.00GiB, used=1.21G
On Tue, Sep 8, 2015 at 11:23 AM, Anna Schumaker
wrote:
> On 09/08/2015 11:21 AM, Pádraig Brady wrote:
>> I see copy_file_range() is a reflink() on BTRFS?
>> That's a bit surprising, as it avoids the copy completely.
>> cp(1) for example considered doing a BTRFS clone by default,
>> but didn't due
On 09/08/2015 11:21 AM, Pádraig Brady wrote:
> On 04/09/15 21:16, Anna Schumaker wrote:
>> Copy system calls came up during Plumbers a couple of weeks ago, because
>> several filesystems (including NFS and XFS) are currently working on copy
>> acceleration implementations. We haven't heard from Za
On 04/09/15 21:16, Anna Schumaker wrote:
> Copy system calls came up during Plumbers a couple of weeks ago, because
> several filesystems (including NFS and XFS) are currently working on copy
> acceleration implementations. We haven't heard from Zach Brown in a while,
> so I volunteered to push hi
On 09/05/2015 04:33 AM, Al Viro wrote:
> On Fri, Sep 04, 2015 at 04:25:27PM -0600, Andreas Dilger wrote:
>
>> This is a bit of a surprising result, since in my testing in the
>> past, copy_{to/from}_user() is a major consumer of CPU time (50%
>> of a CPU core at 1GB/s). What backing filesystem di
On 09/04/2015 06:25 PM, Andreas Dilger wrote:
> On Sep 4, 2015, at 2:16 PM, Anna Schumaker wrote:
>>
>> Copy system calls came up during Plumbers a couple of weeks ago,
>> because several filesystems (including NFS and XFS) are currently
>> working on copy acceleration implementations. We haven't
On 09/04/2015 06:31 PM, Andreas Dilger wrote:
> On Sep 4, 2015, at 3:38 PM, Darrick J. Wong wrote:
>>
>> On Fri, Sep 04, 2015 at 04:17:03PM -0400, Anna Schumaker wrote:
>>> copy_file_range() is a new system call for copying ranges of data
>>> completely in the kernel. This gives filesystems an op
On 09/04/2015 05:38 PM, Darrick J. Wong wrote:
> On Fri, Sep 04, 2015 at 04:17:03PM -0400, Anna Schumaker wrote:
>> copy_file_range() is a new system call for copying ranges of data
>> completely in the kernel. This gives filesystems an opportunity to
>> implement some kind of "copy acceleration",
On 09/04/2015 05:08 PM, Darrick J. Wong wrote:
> On Fri, Sep 04, 2015 at 04:17:02PM -0400, Anna Schumaker wrote:
>> The NFS server will need a fallback for filesystems that don't have any
>> kind of copy acceleration yet, and it should be generally useful to have
>> an in-kernel copy to avoid lots
sequence, transid and reserved field of inode was write to disk
with uninitizlized value, this patch fixed it.
Signed-off-by: Zhao Lei
---
btrfs-convert.c | 4
ctree.h | 2 ++
2 files changed, 6 insertions(+)
diff --git a/btrfs-convert.c b/btrfs-convert.c
index f4fc650..934f4dc 100
We need use ext2fs_close_inode_scan to release relative resource
get from ext2fs_open_inode_scan() in copy_inodes()
Signed-off-by: Zhao Lei
---
btrfs-convert.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/btrfs-convert.c b/btrfs-convert.c
index 934f4dc..e7f3c9e 100644
--- a/btrfs-convert.
s/generation/sequence
for BTRFS_SETGET_STACK_FUNCS(stack_inode_sequence, ...)
Signed-off-by: Zhao Lei
---
ctree.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ctree.h b/ctree.h
index bcad2b9..2061e1e 100644
--- a/ctree.h
+++ b/ctree.h
@@ -1423,7 +1423,7 @@ BTRFS_SETGET_FUN
sequence, transid and reserved field of inode was write to disk
with uninitizlized value, this patchset fixed it, also fixed some
other typo.
Zhao Lei (3):
btrfs-progs: fix set_get typo of btrfs_inode_item->sequence
btrfs-progs: Avoid uninitialized data in output of btrfs-convert
btrfs-progs
On 2015-09-06 13:51, Elliott, Robert (Persistent Memory) wrote:
mkfs.btrfs does not detect pmem devices as being SSDs in kernel 4.2.
Label: (null)
UUID: 46603efe-728c-43fe-8241-ffc125e1a7ed
Node size: 16384
Sector size:4096
Filesystem size:8.00GiB
On Tue, Sep 08, 2015 at 01:59:19PM +0200, Peter Keše wrote:
>
> I'm planning to set up a raid6 array with 4 x 4TB drives.
> Presumably that would result in 8TB of usable space + parity, which
> is about enough for my data (my data is currently 5TB in raid1,
> slowly growing at about 1 TB per year,
I'm planning to set up a raid6 array with 4 x 4TB drives.
Presumably that would result in 8TB of usable space + parity, which is
about enough for my data (my data is currently 5TB in raid1, slowly
growing at about 1 TB per year, but I often keep some additional backups
if space permits).
How
On Sat, Sep 05, 2015 at 01:06:18AM +0200, Arnd Hannemann wrote:
> In newer distros (ubuntu 15.10, fedora rawhide) the binutils
> ar uses the new D flag per default to build deterministic
> binaries.
> Without this patch the following warning is issued, when
> building btrfs-progs:
>
> [AR]
On Tue, Sep 08, 2015 at 05:02:32PM +0800, Sheng Yong wrote:
> Hi, Qu
>
> On 9/8/2015 4:50 PM, Qu Wenruo wrote:
> > Sheng Yong wrote on 2015/09/08 08:46 +:
> >> * If the allocation failed, don't free to free it, even though kfree
> >>allows to free a NULL pointer.
> >> * If posix_acl_to_xat
From: Anand Jain
Controlled EIO from the device is achieved using the dm device.
Helper functions are at common/dmerror.
Broadly steps will include calling _dmerror_init().
_dmerror_init() will use SCRATCH_DEV to create dm linear device and assign
DMERROR_DEV to /dev/mapper/error-test.
When tes
Cleanup the old facilities which use old btrfs_qgroup_reserve() function
call, replace them with the newer version, and remove the "__" prefix in
them.
Also, make btrfs_qgroup_reserve/free() functions private, as they are
now only used inside qgroup codes.
Now, the whole btrfs qgroup is swithed t
Use new __btrfs_delalloc_reserve_space to reserve space.
Signed-off-by: Qu Wenruo
---
fs/btrfs/inode-map.c | 2 +-
fs/btrfs/inode.c | 16 ++--
fs/btrfs/ioctl.c | 5 +++--
3 files changed, 14 insertions(+), 9 deletions(-)
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-m
For NOCOW and inline case, there will be no delayed_ref created for
them, so we should free their reserved data space at proper
time(finish_ordered_io for NOCOW and cow_file_inline for inline).
Signed-off-by: Qu Wenruo
---
fs/btrfs/extent-tree.c | 7 ++-
fs/btrfs/inode.c | 15
Add new version of btrfs_delalloc_reserve_space() function, which
supports accurate qgroup reserve.
Signed-off-by: Qu Wenruo
---
fs/btrfs/ctree.h | 1 +
fs/btrfs/extent-tree.c | 38 ++
2 files changed, 39 insertions(+)
diff --git a/fs/btrfs/ctree.h b/f
As we have the new metadata reservation functions, use them to replace
the old btrfs_qgroup_reserve() call for metadata.
Signed-off-by: Qu Wenruo
---
fs/btrfs/extent-tree.c | 14 ++
fs/btrfs/transaction.c | 34 ++
fs/btrfs/transaction.h | 1 -
3 files
Use new check_data_free_space for buffered write and inode cache.
For buffered write case, as nodatacow write won't increase quota account,
so unlike old behavior which does reserve before check nocow, now we
check nocow first and then only reserve data if we can't do nocow write.
Signed-off-by:
Add new function __btrfs_check_data_free_space() to do precious space
reservation.
The new function will replace old btrfs_check_data_free_space(), but
until all the change is done, let's just use the new name.
Also, export internal use function btrfs_alloc_data_chunk_ondemand(), as
now qgroup re
Now fallocate will do accurate qgroup reserve space check, unlike old
method, which will always reserve the whole length of the range.
With this patch, fallocate will:
1) Iterate the desired range and mark in data rsv map
Only range which is going to be allocated will be recorded in data
rsv
Introduce new functions btrfs_qgroup_reserve/free_meta() to reserve/free
metadata reserved space.
Signed-off-by: Qu Wenruo
---
fs/btrfs/ctree.h | 3 +++
fs/btrfs/disk-io.c | 1 +
fs/btrfs/qgroup.c | 40
fs/btrfs/qgroup.h | 4
4 files changed,
Qgroup reserved space needs to be released from inode dirty map and get
freed at different timing:
1) Release when the metadata is written into tree
After corresponding metadata is written into tree, any newer write will
be COWed(don't include NOCOW case yet).
So we must release its range from ino
Add new function btrfs_add_delayed_qgroup_reserve() function to record
how much space is reserved for that extent.
As btrfs only accounts qgroup at run_delayed_refs() time, so newly
allocated extent should keep the reserved space until then.
So add needed function with related members to do it.
Introduce functions btrfs_qgroup_release/free_data() to release/free
reserved data range.
Release means, just remove the data range from data rsv map, but doesn't
free the reserved space.
This is for normal buffered write case, when data is written into disc
and its metadata is added into tree, it
Sorry for the confusing cover letter title.
This patch is no longer RFC now.
It's already a working one, and we're doing stress test to ensure it's
completely OK, but seems quite good for now.
To Chris,
I know the timing I sent the patchset is quite awful, as there is only
less than 1 week f
Introduce new function reserve_data_range().
This function will find non-overlap range and to insert it into reserve
map using previously introduced functions.
This provides the basis for later per inode reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 92
New function insert_data_ranges() will insert non-overlap reserve ranges
into reserve map.
It provides the basis for later qgroup reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 124 ++
1 file changed, 124 insertions(+)
This new function will do all the hard work to reserve precious space
for a write.
The overall work flow will be the following.
File A already has some dirty pages:
0 4K 8K 12K 16K
|///| |///|
And then, someone want to write some data into range [4K, 16K).
Introduce new function release_data_range() to release reserved ranges.
It will iterate through all existing ranges and remove/shrink them.
Note this function will not free reserved space, as the range can be
released in the following conditions:
1) The dirty range gets written to disk.
In this
Hi, Qu
On 9/8/2015 4:50 PM, Qu Wenruo wrote:
> Sheng Yong wrote on 2015/09/08 08:46 +:
>> * If the allocation failed, don't free to free it, even though kfree
>>allows to free a NULL pointer.
>> * If posix_acl_to_xattr() failed, cleanup the allocation and return
>>the error directly.
>
Introduce the new function to search the most left reserve range in a
reserve map.
It provides the basis for later reserve map implement.
Signed-off-by: Qu Wenruo
---
fs/btrfs/qgroup.c | 36
1 file changed, 36 insertions(+)
diff --git a/fs/btrfs/qgroup.c b/
Add new structures and functions for new qgroup reserve implement dirty
phase.
Which will focus on avoiding over-reserve as in that case, which means
for already reserved dirty space range, we won't reserve space again.
This patch adds the needed structure declaration and comments.
Signed-off-by:
New functions btrfs_qgroup_init/free_data_rsv_map() to init/free data
reserve map.
Data reserve map is used to mark which range already holds reserved
space, to avoid current reserved space leak.
Signed-off-by: Qu Wenruo
---
fs/btrfs/btrfs_inode.h | 2 ++
fs/btrfs/inode.c | 10 +++
f
[[BUG]]
One of the most common case to trigger the bug is the following method:
1) Enable quota
2) Limit excl of qgroup 5 to 16M
3) Write [0,2M) of a file inside subvol 5 10 times without sync
EQUOT will be triggered at about the 8th write.
[[CAUSE]]
The problem is caused by the fact that qgroup
Sheng Yong wrote on 2015/09/08 08:46 +:
* If the allocation failed, don't free to free it, even though kfree
allows to free a NULL pointer.
* If posix_acl_to_xattr() failed, cleanup the allocation and return
the error directly.
So, what's the point?
For me, I didn't see the pros of the
* If the allocation failed, don't free to free it, even though kfree
allows to free a NULL pointer.
* If posix_acl_to_xattr() failed, cleanup the allocation and return
the error directly.
Signed-off-by: Sheng Yong
---
fs/btrfs/acl.c | 14 +++---
1 file changed, 7 insertions(+), 7 del
97 matches
Mail list logo
