Re: [LTP] ffsb failures running ltp-ext4 new features test on SLES11sp1 PPC

[Akira Fujita <a-fujita@rs.jp.nec.com>]

Hi Subrata,

> Akira,
>
> Probably we need to change the tests conditions to restrict not to cross
> the size of partition. We can handle the tests better i think. I am not
> sure if the last commit is playing foul with the tests:
>
> commit 7f20a3add6c7f2166dbab5a712713ae8b5943448
> Author: Subrata Modak<subratamodak@subratamodak.linux.ibm.com>
> Date:   Wed Jun 2 18:16:46 2010 +0530

Yes, we should check ext4 partition size and configure fssb write size
before testing.

AFAIK, ext4-subdir-limit test does not depend on partition size,
and over 32000 directory feature is not supported in case
1KB block size and long directory name.
So I think my fix (just skip invalid test case) is okay.

Reference: linux/Documentation/filesystems/ext4.txt

* lift 32000 subdirectory limit imposed by i_links_count[1]
[1] Filesystems with a block size of 1k may see a limit imposed by the
directory hash tree having a maximum depth of two.

Regards,
Akira Fujita

(2010/06/19 2:30), Subrata Modak wrote:
> Yes, i find many failures in the tests including dumping of BINARY
> characters on this screen. Following are excerpts of the failures:
> 
> ext4-journal-checksum    7  TFAIL  :  ffsb returned failure
> ext4-journal-checksum    0  TINFO  :  journal mode: ordered, commit
> interval: 1,
> ext4-journal-checksum    8  TPASS  :  ext4 journal checksum test pass
> ext4-journal-checksum    0  TINFO  :  journal mode: journal, commit
> interval: 1,
> ext4-journal-checksum    9  TPASS  :  ext4 journal checksum test pass
> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit
> interval: 100,
> write: No space left on device
> write: No space left on device
> write: No space left on device
> \x11]�0��������\x11]�@\x11]����O\x10�����0\x01\x01�碘
> �p\x02A��,�\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x14\x11]\x7fP\x17\x11]�p��������\x11]����N\x10�����0\x01\x01�����p\x02A��$�\x11]��\x11\x11]����������\x11]�������0�p\x02A��$�\x11]��\x12\x11]����������\x11]������0�p\x01w�\x11]}���������\x11]~\x11]\x7fP��\x10�����0\x01\x01�ﾏ��p\x02A��,�\x11]\x7f�\x11]\x7f�\x11]\x7f�\x11]\x7f���O\x11]\x7f�\x11]\x7f�\x11]\x7f�\x11]}\x10\x16\x11]�0��������\x11]�@\x11]����O\x10�����0\x01\x01�碘�p\x02A��,�\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x13\x11]�\x14\x11]\x7fP\x17\x11]�p��������\x11]����N\x10�����0\x01\x01�����p\x02A��$�\x11]��\x11\x11]����������\x11]�������0�p\x02A��$�\x11]��\x12\x11]����������\x11]������0�p\x01w�
> 
> Akira,
> 
> Probably we need to change the tests conditions to restrict not to cross
> the size of partition. We can handle the tests better i think. I am not
> sure if the last commit is playing foul with the tests:
> 
> commit 7f20a3add6c7f2166dbab5a712713ae8b5943448
> Author: Subrata Modak<subratamodak@subratamodak.linux.ibm.com>
> Date:   Wed Jun 2 18:16:46 2010 +0530
> 
> Regards--
> Subrata
> 
> On Fri, 2010-06-18 at 11:39 +0530, Subrata Modak wrote:
>> Hi Lucio,
>>
>> Till now i did not detect any failure for the sdb2:
>>
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>> 0, Blocksize: 1024, isAuto_da_alloc: auto_da_alloc=1
>> ext4-delalloc-mballoc    1  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>> 0, Blocksize: 1024, isAuto_da_alloc: noauto_da_alloc
>> ext4-delalloc-mballoc    2  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: nodelalloc,
>> isDirectIO: 0, Blocksize: 1024, isAuto_da_alloc: auto_da_alloc=1
>> ext4-delalloc-mballoc    3  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: nodelalloc,
>> isDirectIO: 0, Blocksize: 1024, isAuto_da_alloc: noauto_da_alloc
>> ext4-delalloc-mballoc    4  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>> 1, Blocksize: 1024, isAuto_da_alloc: auto_da_alloc=1
>> ext4-delalloc-mballoc    5  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>> 1, Blocksize: 1024, isAuto_da_alloc: noauto_da_alloc
>> ext4-delalloc-mballoc    6  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: nodelalloc,
>> isDirectIO: 1, Blocksize: 1024, isAuto_da_alloc: auto_da_alloc=1
>> ext4-delalloc-mballoc    7  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: nodelalloc,
>> isDirectIO: 1, Blocksize: 1024, isAuto_da_alloc: noauto_da_alloc
>> ext4-delalloc-mballoc    8  TPASS  :  delalloc/mballoc test pass
>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>> 0, Blocksize: 4096, isAuto_da_alloc: auto_da_alloc=1
>>
>> I will continue to run the tests and update you again.
>>
>> Regards--
>> Subrata
>>
>> On Thu, 2010-06-17 at 18:12 -0300, Lucio Correia wrote:
>>> Hi Akira,
>>>
>>> I've found some failures running ext4-newfeat with SLES11sp1 on POWER6.
>>> Have you seen that before?
>>>
>>>
>>>
>>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>>> 0,
>>> Blocksize: 4096, isAuto_da_alloc: auto_da_alloc=1
>>> write: No space left on device
>>> ext4-delalloc-mballoc    9  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 1,
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum    1  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 100,
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum    4  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 1,
>>> write: Success
>>> write: Success
>>> write: No space left on device
>>> ext4-journal-checksum    7  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 100,
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum   10  TFAIL  :  ffsb returned failure
>>>
>>> and so on 13, 16, 19, 22, 25, 28, 31, 34.
>>> ext4-delalloc-mballoc    0  TINFO  :  isDelalloc: delalloc, isDirectIO:
>>> 0,
>>> Blocksize: 4096, isAuto_da_alloc: auto_da_alloc=1
>>> write: No space left on device
>>> ext4-delalloc-mballoc    9  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 1,
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum    1  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 100,
>>> write: No space left on device
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum    4  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 1,
>>> write: Success
>>> write: Success
>>> write: No space left on device
>>> ext4-journal-checksum    7  TFAIL  :  ffsb returned failure
>>>
>>> ext4-journal-checksum    0  TINFO  :  journal mode: writeback, commit interval:
>>> 100,
>>> write: No space left on device
>>> write: No space left on device
>>> ext4-journal-checksum   10  TFAIL  :  ffsb returned failure
>>>
>>> and so on 13, 16, 19, 22, 25, 28, 31, 34.
>>>
>>>
>>> Regards,
>>>
>>>
>>>
>>> Lucio Correia
>>> --------------
>>> Software Engineer
>>> IBM LTC
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>>> lucky parental unit.  See the prize list and enter to win:
>>> http://p.sf.net/sfu/thinkgeek-promo
>>> _______________________________________________
>>> Ltp-list mailing list
>>> Ltp-list@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/ltp-list
>>
>>
>> ------------------------------------------------------------------------------
>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>> lucky parental unit.  See the prize list and enter to win:
>> http://p.sf.net/sfu/thinkgeek-promo
>> _______________________________________________
>> Ltp-list mailing list
>> Ltp-list@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/ltp-list
> 
> 

-- 
Akira Fujita <a-fujita@rs.jp.nec.com>

The First Fundamental Software Development Group,
Software Development Division,
NEC Software Tohoku, Ltd.


------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list