Re: [Testing] Testing Reiser4 with xfstests

Re: [Testing] Testing Reiser4 with xfstests

[Dušan Čolić]

On Fri, Dec 5, 2014 at 2:22 PM, Dušan Čolić <dusanc@gmail.com> wrote:
> I managed to get xfstests  (
> http://oss.sgi.com/cgi-bin/gitweb.cgi?p=xfs/cmds/xfstests.git;a=tree ) to
> run on Reiser4 partition. Test for async, acl, quota still don't work but a
> lot of the rest finishes successfully.
> Tests were run on reiser4-for-3.16.2.patch
> I'll post unsuccessful tests and their results in following e-mails.
>
Btw. these tests can stress test fs and I could also test with
different mount (block size?) or mkfs options (reg40) and automate it
easily to test multiple combinations, so if there's any ideas what
would be useful please tell me.

Thanks
Dushan


> Dushan
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [Testing] Testing Reiser4 with xfstests

[Dušan Čolić]

Some of the tests (~20 of 300) didn't work and I traced that they
require fallocate. AFAIK R4 doesn't support it right?

On Fri, Dec 5, 2014 at 2:46 PM, Dušan Čolić <dusanc@gmail.com> wrote:
> On Fri, Dec 5, 2014 at 2:22 PM, Dušan Čolić <dusanc@gmail.com> wrote:
>> I managed to get xfstests  (
>> http://oss.sgi.com/cgi-bin/gitweb.cgi?p=xfs/cmds/xfstests.git;a=tree ) to
>> run on Reiser4 partition. Test for async, acl, quota still don't work but a
>> lot of the rest finishes successfully.
>> Tests were run on reiser4-for-3.16.2.patch
>> I'll post unsuccessful tests and their results in following e-mails.
>>
> Btw. these tests can stress test fs and I could also test with
> different mount (block size?) or mkfs options (reg40) and automate it
> easily to test multiple combinations, so if there's any ideas what
> would be useful please tell me.
>
> Thanks
> Dushan
>
>
>> Dushan
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [Testing] Testing Reiser4 with xfstests

[Edward Shishkin]

On 12/06/2014 11:07 AM, Dušan Čolić wrote:
> Some of the tests (~20 of 300) didn't work and I traced that they
> require fallocate. AFAIK R4 doesn't support it right?


Yup, we don't support fallocate.

Edward.


>
> On Fri, Dec 5, 2014 at 2:46 PM, Dušan Čolić <dusanc@gmail.com> wrote:
>> On Fri, Dec 5, 2014 at 2:22 PM, Dušan Čolić <dusanc@gmail.com> wrote:
>>> I managed to get xfstests  (
>>> http://oss.sgi.com/cgi-bin/gitweb.cgi?p=xfs/cmds/xfstests.git;a=tree ) to
>>> run on Reiser4 partition. Test for async, acl, quota still don't work but a
>>> lot of the rest finishes successfully.
>>> Tests were run on reiser4-for-3.16.2.patch
>>> I'll post unsuccessful tests and their results in following e-mails.
>>>
>> Btw. these tests can stress test fs and I could also test with
>> different mount (block size?) or mkfs options (reg40) and automate it
>> easily to test multiple combinations, so if there's any ideas what
>> would be useful please tell me.
>>
>> Thanks
>> Dushan
>>
>>
>>> Dushan

--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [Testing] Testing Reiser4 with xfstests

[Dušan Čolić]

On Sat, Dec 6, 2014 at 1:02 PM, Edward Shishkin
<edward.shishkin@gmail.com> wrote:
>
> On 12/06/2014 11:07 AM, Dušan Čolić wrote:
>>
>> Some of the tests (~20 of 300) didn't work and I traced that they
>> require fallocate. AFAIK R4 doesn't support it right?
>
>
>
> Yup, we don't support fallocate.
>
> Edward.
>
>
>
>>
>> On Fri, Dec 5, 2014 at 2:46 PM, Dušan Čolić <dusanc@gmail.com> wrote:
>>>
>>> On Fri, Dec 5, 2014 at 2:22 PM, Dušan Čolić <dusanc@gmail.com> wrote:
>>>>
>>>> I managed to get xfstests  (
>>>> http://oss.sgi.com/cgi-bin/gitweb.cgi?p=xfs/cmds/xfstests.git;a=tree )
>>>> to
>>>> run on Reiser4 partition. Test for async, acl, quota still don't work
>>>> but a
>>>> lot of the rest finishes successfully.
>>>> Tests were run on reiser4-for-3.16.2.patch
>>>> I'll post unsuccessful tests and their results in following e-mails.
>>>>
>>> Btw. these tests can stress test fs and I could also test with
>>> different mount (block size?) or mkfs options (reg40) and automate it
>>> easily to test multiple combinations, so if there's any ideas what
>>> would be useful please tell me.
>>>
All combinations I test fail same tests:

generic/091 fsx exercising direct IO -- sub-block sizes and concurrent
buffered IO
generic/130 SQA Test No. 130

xfs_io vector read/write and trunc tests. modified from cxfsqa tests
 - unixfile_basic_block_hole
- unixfile_buffer_direct_
coherency
 - unixfile_direct_rw
 - unixfile_eof_direct
 - unixfile_fsb_edge
 - unixfile_open_append
 - unixfile_open_trunc
 - unixfile_small_vector_async_rw
 - unixfile_small_vector_sync_rw

generic/135 Concurrent I/O to same file to ensure no deadlocks

generic/263 fsx exercising direct IO vs sub-block buffered I/O


I'm currently testing these combinations (only ccreg40 as reg40 is
stuck in D state in the middle of the tests):

[r4ccHybrid]

MODULAR=0
TEST_DIR=/mnt/test
SCRATCH_MNT=/mnt/scratch
TEST_DEV=/dev/hda5
SCRATCH_DEV=/dev/hda7

RESULT_BASE="`pwd`/results/`date +%d%m%y_%H%M%S`"
RECREATE_TEST_DEV="true"

FSTYP=reiser4
MKFS_OPTIONS=" "
MOUNT_OPTIONS=" "

[r4ccNoatime]
MKFS_OPTIONS=" "
MOUNT_OPTIONS="-o noatime"

[r4ccCOW]
MKFS_OPTIONS=" "
MOUNT_OPTIONS="-o txmod=wa"

[r4ccJournal]
MKFS_OPTIONS=" "
MOUNT_OPTIONS="-o txmod=journal"

[r4ccLexic]
MKFS_OPTIONS="-o fibration=lexic_fibre"
MOUNT_OPTIONS=" "

[r4ccDot_o]
MKFS_OPTIONS="-o fibration=dot_o_fibre"
MOUNT_OPTIONS=" "

[r4ccExt_1]
MKFS_OPTIONS="-o fibration=ext_1_fibre"
MOUNT_OPTIONS=" "

[r4ccExt_3]
MKFS_OPTIONS="-o fibration=ext_3_fibre"
MOUNT_OPTIONS=" "

[r4ccTails]
MKFS_OPTIONS="-o formatting=tails"
MOUNT_OPTIONS=" "

[r4ccExtents]
MKFS_OPTIONS="-o formatting=extents"
MOUNT_OPTIONS=" "


>>> Thanks
>>> Dushan
>>>
>>>
>>>> Dushan
>
>
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html