On Tue, May 14, 2013 at 11:36:53AM +0200, Stefan Behrens wrote:
> Mapping UUIDs to subvolume IDs is an operation with a high effort
> today. Today, the algorithm even has quadratic effort (based on the
> number of existing subvolumes), which means, that it takes minutes
> to send/receive a single s
On Thu, May 16, 2013 at 01:34:11PM +0800, Miao Xie wrote:
> On Thu, 16 May 2013 13:15:57 +0800, Liu Bo wrote:
> > On Thu, May 16, 2013 at 12:31:11PM +0800, Miao Xie wrote:
> >> On thu, 16 May 2013 11:36:46 +0800, Liu Bo wrote:
> >>> On Wed, May 15, 2013 at 03:48:20PM +0800, Miao Xie wrote:
> T
On Thu, 16 May 2013 13:15:57 +0800, Liu Bo wrote:
> On Thu, May 16, 2013 at 12:31:11PM +0800, Miao Xie wrote:
>> On thu, 16 May 2013 11:36:46 +0800, Liu Bo wrote:
>>> On Wed, May 15, 2013 at 03:48:20PM +0800, Miao Xie wrote:
The grab/put funtions will be used in the next patch, which need grab
On Thu, May 16, 2013 at 12:31:11PM +0800, Miao Xie wrote:
> On thu, 16 May 2013 11:36:46 +0800, Liu Bo wrote:
> > On Wed, May 15, 2013 at 03:48:20PM +0800, Miao Xie wrote:
> >> The grab/put funtions will be used in the next patch, which need grab
> >> the root object and ensure it is not freed. We
On thu, 16 May 2013 11:36:46 +0800, Liu Bo wrote:
> On Wed, May 15, 2013 at 03:48:20PM +0800, Miao Xie wrote:
>> The grab/put funtions will be used in the next patch, which need grab
>> the root object and ensure it is not freed. We use reference counter
>> instead of the srcu lock is to aovid bloc
On Thu, 16 May 2013 11:20:39 +0800, Liu Bo wrote:
> On Wed, May 15, 2013 at 03:48:24PM +0800, Miao Xie wrote:
>> Before applying this patch, we need flush all the delalloc inodes in
>> the fs when we want to create a snapshot, it wastes time, and make
>> the transaction commit be blocked for a long
On Wed, May 15, 2013 at 03:48:20PM +0800, Miao Xie wrote:
> The grab/put funtions will be used in the next patch, which need grab
> the root object and ensure it is not freed. We use reference counter
> instead of the srcu lock is to aovid blocking the memory reclaim task,
> which invokes synchroni
On Wed, May 15, 2013 at 03:48:24PM +0800, Miao Xie wrote:
> Before applying this patch, we need flush all the delalloc inodes in
> the fs when we want to create a snapshot, it wastes time, and make
> the transaction commit be blocked for a long time. It means some other
> user operation would also
On Wed, May 15, 2013 at 05:39:35PM +0200, Stefan Behrens wrote:
> On Tue, 14 May 2013 18:44:11 +0800, Liu Bo wrote:
> > On Tue, May 14, 2013 at 11:36:56AM +0200, Stefan Behrens wrote:
> >> @@ -396,7 +403,7 @@ static noinline int create_subvol(struct inode *dir,
> >> * of create_snapshot().
> >>
Current way of specifying the path to match is not very comfortable, but
the feature itself is very useful. Let's save the short option -m for a
more user friendly syntax and keep a long option --path-regex with the
current syntax.
CC: Peter Stuge
Signed-off-by: David Sterba
---
cmds-restore.c
On Wed, May 15, 2013 at 08:21:54PM +, Myklebust, Trond wrote:
> On Wed, 2013-05-15 at 16:19 -0400, J. Bruce Fields wrote:
> > On Tue, May 14, 2013 at 02:15:26PM -0700, Zach Brown wrote:
> > > This crude patch illustrates the simplest plumbing involved in
> > > supporting sys_call_range with the
On Wed, 2013-05-15 at 16:19 -0400, J. Bruce Fields wrote:
> On Tue, May 14, 2013 at 02:15:26PM -0700, Zach Brown wrote:
> > This crude patch illustrates the simplest plumbing involved in
> > supporting sys_call_range with the NFS COPY operation that's pending in
> > the 4.2 draft spec.
> >
> > The
On Tue, May 14, 2013 at 02:15:26PM -0700, Zach Brown wrote:
> This crude patch illustrates the simplest plumbing involved in
> supporting sys_call_range with the NFS COPY operation that's pending in
> the 4.2 draft spec.
>
> The patch is based on a previous prototype that used the COPY op to
> imp
On Wed, May 15, 2013 at 07:44:05PM +, Eric Wong wrote:
> Why introduce a new syscall instead of extending sys_splice?
Personally, I think it's ugly to have different operations use the same
syscall just because their arguments match.
But that preference aside, sure, if the consensus is that w
Zach Brown wrote:
> This adds a syscall and vfs entry point for clone_range which offloads
> data copying between existing files.
>
> The syscall is a thin wrapper around the vfs entry point. Its arguments
> are inspired by sys_splice().
Why introduce a new syscall instead of extending sys_spli
On May 15, 2013, Josef Bacik wrote:
> So this should only happen in the case that you are on a dm device it looks
> like, is that how you are running?
That was my first thought, but no, I'm using partitions out of the SATA
disks directly. I even checked for stray dm out of fake raid or
somesuch
On May 14, 2013, Liu Bo wrote:
>> In one of the failures that caused machine load spikes, I tried to
>> collect info on active processes with perf top and SysRq-T, but nothing
>> there seemed to explain the spike. Thoughts on how to figure out what's
>> causing this?
> Although I've seen your s
Left out a newline in the generation check printf.
Signed-off-by: Josef Bacik
---
free-space-cache.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/free-space-cache.c b/free-space-cache.c
index 5fb8ece..a30438b 100644
--- a/free-space-cache.c
+++ b/free-space-cache.c
@
On Wed, May 15, 2013 at 7:28 PM, Chris Murphy wrote:
>
> On May 15, 2013, at 10:44 AM, Harald Glatt wrote:
>
>>
>> You make a ro snapshot rw by creating a snapshot of it that is rw. So
>> yes to both questions, by doing the same thing in both cases.
>
> In other words, a normal snapshot (without
On May 15, 2013, at 10:44 AM, Harald Glatt wrote:
>
> You make a ro snapshot rw by creating a snapshot of it that is rw. So
> yes to both questions, by doing the same thing in both cases.
In other words, a normal snapshot (without -r) of a read-only snapshot will
create a rw snapshot?
In any
On Wed, May 15, 2013 at 6:43 PM, Chris Murphy wrote:
>
> On May 15, 2013, at 1:40 AM, Gabriel de Perthuis wrote:
>
>>
>> You can move subvolumes at any time, as if they were regular directories.
>
> In the example case, the subvolumes are read-only. So is it possible to make
> a read-only subvol
On May 15, 2013, at 1:40 AM, Gabriel de Perthuis wrote:
>
> You can move subvolumes at any time, as if they were regular directories.
In the example case, the subvolumes are read-only. So is it possible to make a
read-only subvolume (snapshot) read-writable? And is it possible to make a
read
On Tue, 14 May 2013 18:44:11 +0800, Liu Bo wrote:
> On Tue, May 14, 2013 at 11:36:56AM +0200, Stefan Behrens wrote:
>> @@ -396,7 +403,7 @@ static noinline int create_subvol(struct inode *dir,
>> * of create_snapshot().
>> */
>> ret = btrfs_subvolume_reserve_metadata(root, &block_rs
On 15 May 2013 17:00, Harald Glatt wrote:
> On Wed, May 15, 2013 at 4:19 PM, Philipp Dreimann
> wrote:
>> Hello,
>>
>> my btrfs filesystem was not mountable anymore after a loss of power:
>>
>> kernel BUG at fs/btrfs/free-space-cache.c:1567!
>> invalid opcode: [#1] SMP
>> Modules linked in:
On Wed, May 15, 2013 at 4:19 PM, Philipp Dreimann wrote:
> Hello,
>
> my btrfs filesystem was not mountable anymore after a loss of power:
>
> kernel BUG at fs/btrfs/free-space-cache.c:1567!
> invalid opcode: [#1] SMP
> Modules linked in: btrfs libcrc32c xor zlib_deflate raid6_pq i915(+)
> i2
On Tue, May 14, 2013 at 07:50:28PM +0800, Wang Shilong wrote:
> When making btrfs filesystem. we firstly write root leaf to
> specified filed, and then we recow the root. If we don't recow,
> some trees are not in the correct block group.
>
> Steps to reproduce:
> dd if=/dev/zero of=test.img
Hello,
my btrfs filesystem was not mountable anymore after a loss of power:
kernel BUG at fs/btrfs/free-space-cache.c:1567!
invalid opcode: [#1] SMP
Modules linked in: btrfs libcrc32c xor zlib_deflate raid6_pq i915(+)
i2c_algo_bit drm_kms_helper drm i2c_core video uinput
CPU 3
Pid: 147, comm
On Fri, May 03, 2013 at 08:56:54AM -0400, Josef Bacik wrote:
> Apparently 512mb was too small, with a fs_mark command we could get so much
> delayed work built up that we'd never trip the "lets commit the transaction"
> logic until we'd gotten too much delayed refs built up. Increasing this to 1
>
Check whether any involved device is already busy running a
scrub. This would cause damaged status messages and the state
"aborted" without the explanation that a scrub was already
running. Therefore check it first, prevent it and give some
feedback to the user if scrub is already running.
Note tha
On Sat, May 11, 2013 at 01:16:38AM -0600, Alexandre Oliva wrote:
> On Apr 4, 2013, Alexandre Oliva wrote:
>
> > I've been trying to figure out the btrfs I/O stack to try to understand
> > why, sometimes (but not always), after a failure to read a (data
> > non-replicated) block from the disk, th
On Tue, May 14, 2013 at 07:11:54PM +0200, Stefan Behrens wrote:
> # of subvols | without| with
> in filesystem | UUID tree | UUID tree
> --++--
> 2 | 0m00.004s | 0m00.003s
> 1000 | 0m07.010s | 0m00.004s
> 2000 | 0m28.210s | 0m00
On 04/26/2013 01:45 PM, Eric Sandeen wrote:
test 284 had... some issues.
diff --git a/tests/btrfs/284 b/tests/btrfs/284
old mode 100644
new mode 100755
index d952977..67161a3
--- a/tests/btrfs/284
+++ b/tests/btrfs/284
This patch has been committed:
commit 91f87e3b89c0f7350a56d397ba7255
On Fri, May 03, 2013 at 11:28:35PM +0800, Liu Bo wrote:
> On Fri, May 03, 2013 at 08:56:54AM -0400, Josef Bacik wrote:
> > Apparently 512mb was too small, with a fs_mark command we could get so much
> > delayed work built up that we'd never trip the "lets commit the transaction"
> > logic until we'
On Tue, 14 May 2013 19:11:54 +0200, Stefan Behrens wrote:
> On Tue, 14 May 2013 18:55:23 +0800, Liu Bo wrote:
>> On Tue, May 14, 2013 at 11:36:52AM +0200, Stefan Behrens wrote:
>>> Mapping UUIDs to subvolume IDs is an operation with a high effort
>>> today. Today, the algorithm even has quadratic e
This patchset improve the problem that the transaction may be blocked for a
long time when it is being committed if there is heavy I/O.
In this patchset,
- 0001-0005, 0007, 0011-0012 are random fix or code cleanup patch.
- 0006, 0008-0010 introduce per-subvolume delalloc inode list and ordered
e
The grab/put funtions will be used in the next patch, which need grab
the root object and ensure it is not freed. We use reference counter
instead of the srcu lock is to aovid blocking the memory reclaim task,
which invokes synchronize_srcu().
Signed-off-by: Miao Xie
---
fs/btrfs/ctree.h |
We used ->num_joined track if there were some writers which join the current
transaction when the committer was sleeping. If some writers joined the current
transaction, we has to continue the while loop to do some necessary stuff, such
as flush the ordered operations. But it is unnecessary because
When we umount a fs with serious errors, we will invoke
btrfs_cleanup_transactions()
to clean up the residual transaction. At this time, It is impossible to start a
new
transaction, so we needn't assign trans_no_join to 1, and also needn't clear
running
transaction every time we destroy a residu
We have checked if ->node is NULL or not, so it is unnecessary to
use BUG_ON() to check again. Remove it.
Signed-off-by: Miao Xie
---
fs/btrfs/disk-io.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 2a9ae38..8c1e4fb 100644
--- a/fs/btrfs/disk-io
The reason we introduce per-subvolume ordered extent list is the same
as the per-subvolume delalloc inode list.
Signed-off-by: Miao Xie
---
fs/btrfs/ctree.h| 25 ---
fs/btrfs/dev-replace.c | 4 +-
fs/btrfs/disk-io.c | 45 +++-
fs/btrfs/extent-tree.c |
Signed-off-by: Miao Xie
---
fs/btrfs/super.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index a4807ce..f0857e0 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1263,6 +1263,7 @@ static int btrfs_remount(struct super_block *sb, int
*flags, char
Before applying this patch, we need flush all the delalloc inodes in
the fs when we want to create a snapshot, it wastes time, and make
the transaction commit be blocked for a long time. It means some other
user operation would also be blocked for a long time.
This patch improves this problem, we
If the transaction is removed from the transaction list, it means the
transaction has been committed successfully. So it is impossible to
call cleanup_transaction(), otherwise there is something wrong with
the code logic. Thus, we use BUG_ON() instead of the original handle.
Signed-off-by: Miao Xi
We used 3 variants to track the state of the transaction, it was complex
and wasted the memory space. Besides that, it was hard to understand that
which types of the transaction handles should be blocked in each transaction
state, so the developers often made mistakes.
This patch improved the abov
We checked the commit time to avoid committing the transaction
frequently, but it is unnecessary because:
- It made the transaction commit spend more time, and delayed the
operation of the external writers(TRANS_START/TRANS_USERSPACE).
- Except the space that we have to commit transaction, such a
btrfs_commit_transaction has the following loop before we commit the
transaction.
do {
// attempt to do some useful stuff and/or sleep
} while (atomic_read(&cur_trans->num_writers) > 1 ||
(should_grow && cur_trans->num_joined != joined));
This is used to prevent from the TRANS_START
There are several functions whose code is similar, such as
btrfs_find_last_root()
btrfs_read_fs_root_no_radix()
Besides that, some functions are invoked twice, it is unnecessary,
for example, we are sure that all roots which is found in
btrfs_find_orphan_roots()
have their orphan items, so i
It is unnecessary to flush the delalloc inodes again and again because
we don't care the dirty pages which are introduced after the flush, and
they will be flush in the transaction commit.
Signed-off-by: Miao Xie
---
fs/btrfs/transaction.c | 26 ++
1 file changed, 18 inse
When we create a snapshot, we need flush all delalloc inodes in the
fs, just flushing the inodes in the source tree is OK. So we introduce
per-subvolume delalloc inode list.
Signed-off-by: Miao Xie
---
fs/btrfs/ctree.h | 22 ---
fs/btrfs/dev-replace.c | 2 +-
fs/btrfs/disk-io.c
btrfs_invalidate_inodes() may sleep, so we should not invoke it in the
spin lock context. Fix it.
Signed-off-by: Miao Xie
---
fs/btrfs/disk-io.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 642c861..724a0da 100644
--- a/fs/btrfs/disk-io.
The root node of the rb-tree may be changed, so we should get it under
the lock. Fix it.
Signed-off-by: Miao Xie
---
fs/btrfs/inode.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 7f6e78a..bf5c399 100644
--- a/fs/btrfs/inode.
inode_tree_del() will move the tree root into the dead root list, and
then the tree will be destroyed by the cleaner. So if we remove the
delayed node which is cached in the inode after inode_tree_del(),
we may access a freed tree root. Fix it.
Signed-off-by: Miao Xie
---
fs/btrfs/inode.c | 2 +-
> A user of a workstation has a home directory /home/john as a subvolume. I
> wrote a cron job to make read-only snapshots of it under /home/john/backup
> which was fortunate as they just ran a script that did something like
> "rm -rf ~".
>
> Apart from copying dozens of gigs of data back, is t
53 matches
Mail list logo
