[PATH]Fix cow semantic in run_delalloc_nocow()

[PATH]Fix cow semantic in run_delalloc_nocow()

[Liu Hui]

Hi,
I found cow doesn't behave as expected in run_delalloc_nocow(). Now in
run_delalloc_now(), if it found a regular extent and the NODATACOW is
set, run_dealloc_now will cow the file range which is not as expected.
Also, if it find a regular extent and NODATACOW is not set(with
PREALLOC set), it will not take cow operation. So, the cow semantic is
opposite to the btrfs design.

I found the problem was introduced by fallocate(), so could Yan please
verify the fix? Much appreciated!

-- 
Thanks & Best Regards
Liu Hui
--

diff --git a/inode.c b/inode.c
index 2c77e09..5525594 100644
--- a/inode.c
+++ b/inode.c
@@ -1003,7 +1003,7 @@ next_slot:
                                goto out_check;
                        if (disk_bytenr == 0)
                                goto out_check;
-                       if (extent_type == BTRFS_FILE_EXTENT_REG && !force)
+                       if (extent_type == BTRFS_FILE_EXTENT_REG && force)
                                goto out_check;
                        if (btrfs_cross_ref_exist(trans, root, disk_bytenr))
                                goto out_check;

Re: [PATH]Fix cow semantic in run_delalloc_nocow()

[Yan Zheng]

2008/11/20 Liu Hui <onlyflyer@gmail.com>:
> Hi,
> I found cow doesn't behave as expected in run_delalloc_nocow(). Now in
> run_delalloc_now(), if it found a regular extent and the NODATACOW is
> set, run_dealloc_now will cow the file range which is not as expected.
> Also, if it find a regular extent and NODATACOW is not set(with
> PREALLOC set), it will not take cow operation. So, the cow semantic is
> opposite to the btrfs design.
>
> I found the problem was introduced by fallocate(), so could Yan please
> verify the fix? Much appreciated!
>
> --
> Thanks & Best Regards
> Liu Hui
> --
>
> diff --git a/inode.c b/inode.c
> index 2c77e09..5525594 100644
> --- a/inode.c
> +++ b/inode.c
> @@ -1003,7 +1003,7 @@ next_slot:
>                                goto out_check;
>                        if (disk_bytenr == 0)
>                                goto out_check;
> -                       if (extent_type == BTRFS_FILE_EXTENT_REG && !force)
> +                       if (extent_type == BTRFS_FILE_EXTENT_REG && force)
>                                goto out_check;
>                        if (btrfs_cross_ref_exist(trans, root, disk_bytenr))
>                                goto out_check;
> --

You are right, thanks. I suggest modifying run_delalloc_range instead of here,
could you please send a new patch.

Yan Zheng

Re: [PATH]Fix cow semantic in run_delalloc_nocow()

[Liu Hui]

Ok, this is the new patch.

-- 
Thanks & Best Regards
Liu Hui
--

diff --git a/inode.c b/inode.c
index 2c77e09..5e3789f 100644
--- a/inode.c
+++ b/inode.c
@@ -1114,10 +1114,10 @@ static int run_delalloc_range(struct inode
*inode, struct page *locked_page,
        if (btrfs_test_opt(root, NODATACOW) ||
            btrfs_test_flag(inode, NODATACOW))
                ret = run_delalloc_nocow(inode, locked_page, start, end,
-                                        page_started, 0, nr_written);
+                                        page_started, 1, nr_written);
        else if (btrfs_test_flag(inode, PREALLOC))
                ret = run_delalloc_nocow(inode, locked_page, start, end,
-                                        page_started, 1, nr_written);
+                                        page_started, 0, nr_written);
        else
                ret = cow_file_range_async(inode, locked_page, start, end,
                                     page_started, nr_written);

Re: [PATH]Fix cow semantic in run_delalloc_nocow()

[Chris Mason]

On Thu, 2008-11-20 at 21:47 +0800, Liu Hui wrote:
> Ok, this is the new patch.
> 

Thank you both, I've pushed this out.

-chris

Re: [PATCH]Fix cow semantic in run_delalloc_nocow()

[Liu Hui]

Hi chris,
This is just a reminder to tell you that I found the commit message in
git log, but the code has not been checked in yet.

2008/11/21 Chris Mason <chris.mason@oracle.com>:
> On Thu, 2008-11-20 at 21:47 +0800, Liu Hui wrote:
>> Ok, this is the new patch.
>>
>
> Thank you both, I've pushed this out.
>
> -chris
>
>
>

-- 
Thanks & Best Regards
Liu Hui
--

Re: [PATCH]Fix cow semantic in run_delalloc_nocow()

[Chris Mason]

On Fri, 2008-11-21 at 12:47 +0800, Liu Hui wrote:
> Hi chris,
> This is just a reminder to tell you that I found the commit message in
> git log, but the code has not been checked in yet.
> 

You're right, somehow the  commit ended up empty.  I'll have to fix this
after vacation.  Thanks.

-chris