diff mbox series

common/btrfs: source module file

Message ID 20211104203958.2371523-1-mcgrof@kernel.org (mailing list archive)
State New, archived
Headers show
Series common/btrfs: source module file | expand

Commit Message

Luis Chamberlain Nov. 4, 2021, 8:39 p.m. UTC
btrfs/249 fails with:

QA output created by 249
./common/btrfs: line 425: _require_loadable_fs_module: command not found
./common/btrfs: line 432: _reload_fs_module: command not found
ERROR: not a btrfs filesystem: /media/scratch

Fix this by sourcing common/module in the btrfs common file.

Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
---
 common/btrfs | 2 ++
 1 file changed, 2 insertions(+)

Comments

Filipe Manana Nov. 5, 2021, 11:04 a.m. UTC | #1
On Thu, Nov 4, 2021 at 8:40 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>
> btrfs/249 fails with:
>
> QA output created by 249
> ./common/btrfs: line 425: _require_loadable_fs_module: command not found
> ./common/btrfs: line 432: _reload_fs_module: command not found
> ERROR: not a btrfs filesystem: /media/scratch
>
> Fix this by sourcing common/module in the btrfs common file.

I'm not sure why you get such a failure. Without the relevant
btrfs-progs and btrfs kernel patches, I don't get that error:

$ ./check btrfs/249
FSTYP         -- btrfs
PLATFORM      -- Linux/x86_64 debian9 5.15.0-rc7-btrfs-next-103 #1 SMP
PREEMPT Tue Nov 2 12:25:45 WET 2021
MKFS_OPTIONS  -- /dev/sdb
MOUNT_OPTIONS -- /dev/sdb /home/fdmanana/btrfs-tests/scratch_1

btrfs/249 [failed, exit status 1]- output mismatch (see
/home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad)
    --- tests/btrfs/249.out 2021-10-26 11:04:03.879678608 +0100
    +++ /home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad
2021-11-05 10:51:53.752113924 +0000
    @@ -1,2 +1,5 @@
     QA output created by 249
    -Silence is golden
    +ERROR: unexpected number of devices: 1 >= 1
    +ERROR: if seed device is used, try running this command as root
    +FAILED: btrfs filesystem usage, ret 1. Check btrfs.ko and
btrfs-progs version.
    +(see /home/fdmanana/git/hub/xfstests/results//btrfs/249.full for details)
    ...
    (Run 'diff -u /home/fdmanana/git/hub/xfstests/tests/btrfs/249.out
/home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad'  to see
the entire diff)
Ran: btrfs/249
Failures: btrfs/249
Failed 1 of 1 tests

Maybe Anand, who authored the test, may have an idea.
We do have many other tests that call
_require_btrfs_forget_or_module_loadable(), btrfs/124, 125, 163, 164,
etc. Does it happen with those as well?

Also, in the future please CC linux-btrfs for changes related to btrfs tests.

Thanks.

>
> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
> ---
>  common/btrfs | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/common/btrfs b/common/btrfs
> index 5d938c19..4dc4f75d 100644
> --- a/common/btrfs
> +++ b/common/btrfs
> @@ -2,6 +2,8 @@
>  # Common btrfs specific functions
>  #
>
> +source common/module
> +
>  _btrfs_get_subvolid()
>  {
>         mnt=$1
> --
> 2.33.0
>
Nikolay Borisov Nov. 5, 2021, 11:32 a.m. UTC | #2
On 5.11.21 г. 13:04, Filipe Manana wrote:
> On Thu, Nov 4, 2021 at 8:40 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>>
>> btrfs/249 fails with:
>>
>> QA output created by 249
>> ./common/btrfs: line 425: _require_loadable_fs_module: command not found
>> ./common/btrfs: line 432: _reload_fs_module: command not found
>> ERROR: not a btrfs filesystem: /media/scratch
>>
>> Fix this by sourcing common/module in the btrfs common file.
> 
> I'm not sure why you get such a failure. Without the relevant
> btrfs-progs and btrfs kernel patches, I don't get that error:

I checked all these tests and btrfs/248 and btrfs/249 do not import
common/module whilst the others do it so this might be one of the reason.

IMO it would be cleaner to source it in common/btrfs to remove the
duplication.


> 
> $ ./check btrfs/249
> FSTYP         -- btrfs
> PLATFORM      -- Linux/x86_64 debian9 5.15.0-rc7-btrfs-next-103 #1 SMP
> PREEMPT Tue Nov 2 12:25:45 WET 2021
> MKFS_OPTIONS  -- /dev/sdb
> MOUNT_OPTIONS -- /dev/sdb /home/fdmanana/btrfs-tests/scratch_1
> 
> btrfs/249 [failed, exit status 1]- output mismatch (see
> /home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad)
>     --- tests/btrfs/249.out 2021-10-26 11:04:03.879678608 +0100
>     +++ /home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad
> 2021-11-05 10:51:53.752113924 +0000
>     @@ -1,2 +1,5 @@
>      QA output created by 249
>     -Silence is golden
>     +ERROR: unexpected number of devices: 1 >= 1
>     +ERROR: if seed device is used, try running this command as root
>     +FAILED: btrfs filesystem usage, ret 1. Check btrfs.ko and
> btrfs-progs version.
>     +(see /home/fdmanana/git/hub/xfstests/results//btrfs/249.full for details)
>     ...
>     (Run 'diff -u /home/fdmanana/git/hub/xfstests/tests/btrfs/249.out
> /home/fdmanana/git/hub/xfstests/results//btrfs/249.out.bad'  to see
> the entire diff)
> Ran: btrfs/249
> Failures: btrfs/249
> Failed 1 of 1 tests
> 
> Maybe Anand, who authored the test, may have an idea.
> We do have many other tests that call
> _require_btrfs_forget_or_module_loadable(), btrfs/124, 125, 163, 164,
> etc. Does it happen with those as well?
> 
> Also, in the future please CC linux-btrfs for changes related to btrfs tests.
> 
> Thanks.
> 
>>
>> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
>> ---
>>  common/btrfs | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/common/btrfs b/common/btrfs
>> index 5d938c19..4dc4f75d 100644
>> --- a/common/btrfs
>> +++ b/common/btrfs
>> @@ -2,6 +2,8 @@
>>  # Common btrfs specific functions
>>  #
>>
>> +source common/module
>> +
>>  _btrfs_get_subvolid()
>>  {
>>         mnt=$1
>> --
>> 2.33.0
>>
> 
>
Luis Chamberlain Nov. 5, 2021, 3:51 p.m. UTC | #3
On Fri, Nov 05, 2021 at 11:04:46AM +0000, Filipe Manana wrote:
> On Thu, Nov 4, 2021 at 8:40 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
> >
> > btrfs/249 fails with:
> >
> > QA output created by 249
> > ./common/btrfs: line 425: _require_loadable_fs_module: command not found
> > ./common/btrfs: line 432: _reload_fs_module: command not found
> > ERROR: not a btrfs filesystem: /media/scratch
> >
> > Fix this by sourcing common/module in the btrfs common file.
> 
> I'm not sure why you get such a failure. Without the relevant
> btrfs-progs and btrfs kernel patches, I don't get that error:

Are you using a latest fstests?

> Maybe Anand, who authored the test, may have an idea.
> We do have many other tests that call
> _require_btrfs_forget_or_module_loadable(), btrfs/124, 125, 163, 164,
> etc. Does it happen with those as well?

If that was not an issue before, maybe it is because of
commit b47ea0c6973 ("common/module: use patient module removal").

> Also, in the future please CC linux-btrfs for changes related to btrfs tests.

Will do, thanks!

 Luis
Luis Chamberlain Nov. 5, 2021, 3:54 p.m. UTC | #4
On Fri, Nov 05, 2021 at 01:32:11PM +0200, Nikolay Borisov wrote:
> 
> 
> On 5.11.21 г. 13:04, Filipe Manana wrote:
> > On Thu, Nov 4, 2021 at 8:40 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
> >>
> >> btrfs/249 fails with:
> >>
> >> QA output created by 249
> >> ./common/btrfs: line 425: _require_loadable_fs_module: command not found
> >> ./common/btrfs: line 432: _reload_fs_module: command not found
> >> ERROR: not a btrfs filesystem: /media/scratch
> >>
> >> Fix this by sourcing common/module in the btrfs common file.
> > 
> > I'm not sure why you get such a failure. Without the relevant
> > btrfs-progs and btrfs kernel patches, I don't get that error:
> 
> I checked all these tests and btrfs/248 and btrfs/249 do not import
> common/module whilst the others do it so this might be one of the reason.

Odd btrfs/248 runs fine here.

> IMO it would be cleaner to source it in common/btrfs to remove the
> duplication.

Sure that works with me. Will send a v2.

  Luis
Anand Jain Nov. 7, 2021, 2:18 p.m. UTC | #5
On 05/11/2021 23:54, Luis Chamberlain wrote:
> On Fri, Nov 05, 2021 at 01:32:11PM +0200, Nikolay Borisov wrote:
>>
>>
>> On 5.11.21 г. 13:04, Filipe Manana wrote:
>>> On Thu, Nov 4, 2021 at 8:40 PM Luis Chamberlain <mcgrof@kernel.org> wrote:
>>>>
>>>> btrfs/249 fails with:
>>>>
>>>> QA output created by 249
>>>> ./common/btrfs: line 425: _require_loadable_fs_module: command not found
>>>> ./common/btrfs: line 432: _reload_fs_module: command not found
>>>> ERROR: not a btrfs filesystem: /media/scratch
>>>>
>>>> Fix this by sourcing common/module in the btrfs common file.
>>>
>>> I'm not sure why you get such a failure. Without the relevant
>>> btrfs-progs and btrfs kernel patches, I don't get that error:
>>
>> I checked all these tests and btrfs/248 and btrfs/249 do not import
>> common/module whilst the others do it so this might be one of the reason.
> 
> Odd btrfs/248 runs fine here.
> 

  Thanks for the report and the fix.

  I still can't find the reason why btrfs/249 runs fine on my test 
system here. (It is the same error as shown in Filipe email).

>> IMO it would be cleaner to source it in common/btrfs to remove the
>> duplication.
> 

  Anyways, in general adding common/module to common/btrfs is a good 
cleanup.

- Anand


> Sure that works with me. Will send a v2.
> 
>    Luis
>
diff mbox series

Patch

diff --git a/common/btrfs b/common/btrfs
index 5d938c19..4dc4f75d 100644
--- a/common/btrfs
+++ b/common/btrfs
@@ -2,6 +2,8 @@ 
 # Common btrfs specific functions
 #
 
+source common/module
+
 _btrfs_get_subvolid()
 {
 	mnt=$1