[LTP] problem : please help !!!

[LTP] problem : please help !!!

[rahul agarwal]

[-- Attachment #1.1: Type: text/plain, Size: 419 bytes --]

I am running the ltp suite on my embedded system and the following two tests
are failing :-


falloc01 , falloc02 , falloac03   ->   TFAIL : TEST_ERRNO=EFBIG(27): File
too large

clone02, clone03, clone05, clone06  ->  TBROK : Unexpected signal 11
received


I have tried everything to figure out the cause of these test failure but I
still couldn't get the problem solved .


With Regards
Rahul Agarwal
+91-9999721591

[-- Attachment #1.2: Type: text/html, Size: 606 bytes --]

[-- Attachment #2: Type: text/plain, Size: 345 bytes --]

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev

[-- Attachment #3: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] problem : please help !!!

[Henry Yei]

On 04/16/2010 04:22 AM, rahul agarwal wrote:
> I am running the ltp suite on my embedded system and the following two tests
> are failing :-
> 
> 
> falloc01 , falloc02 , falloac03   ->   TFAIL : TEST_ERRNO=EFBIG(27): File
> too large
> 
> clone02, clone03, clone05, clone06  ->  TBROK : Unexpected signal 11
> received
> 
> 
> I have tried everything to figure out the cause of these test failure but I
> still couldn't get the problem solved .
> 
> 
> With Regards
> Rahul Agarwal
> +91-9999721591

I assume you mean fallocate01, 02, 03, etc. Are you running these on
ext4 or other filesystem that supports the fallocate calls
(preallocation feature)?

Your architecture may also hit a problem with the tests.


Here are some notes I had with the fallocate tests from a few months
back that may interest you:

On a couple of architectures (arm, mips) that the original authors never
tested on, fallocate calls are returning EFBIG (file too large) rather
than EOPNOTSUPP(not supported) like they are on pcc, and x86 architectures.

I've run it on x86_64 Ubuntu 9.04 with an ext4 system, and the tests
also seem to fail with EFBIG, when they should pass AFAICT.


-- 
Henry Yei <hyei@mvista.com>

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] problem : please help !!!

[Garrett Cooper]

On Mon, Apr 19, 2010 at 2:27 PM, Henry Yei <hyei@mvista.com> wrote:
> On 04/16/2010 04:22 AM, rahul agarwal wrote:
>> I am running the ltp suite on my embedded system and the following two tests
>> are failing :-
>>
>>
>> falloc01 , falloc02 , falloac03   ->   TFAIL : TEST_ERRNO=EFBIG(27): File
>> too large
>>
>> clone02, clone03, clone05, clone06  ->  TBROK : Unexpected signal 11
>> received
>>
>>
>> I have tried everything to figure out the cause of these test failure but I
>> still couldn't get the problem solved .
>
> I assume you mean fallocate01, 02, 03, etc. Are you running these on
> ext4 or other filesystem that supports the fallocate calls
> (preallocation feature)?
>
> Your architecture may also hit a problem with the tests.
>
>
> Here are some notes I had with the fallocate tests from a few months
> back that may interest you:
>
> On a couple of architectures (arm, mips) that the original authors never
> tested on, fallocate calls are returning EFBIG (file too large) rather
> than EOPNOTSUPP(not supported) like they are on pcc, and x86 architectures.
>
> I've run it on x86_64 Ubuntu 9.04 with an ext4 system, and the tests
> also seem to fail with EFBIG, when they should pass AFAICT.

Rahul,
    I never ran into this issue on i686, mips*, ppc, or x86_64 when I
used to run these tests to validate releases, but I haven't actively
run LTP for several months, and my environment most likely varied from
your's.
    For starters, read the note about EFBIG for fallocate(2):
http://www.kernel.org/doc/man-pages/online/pages/man2/fallocate.2.html
. After you've done that, do some additional research to solve your
problem:

1. Consider where the file is being is being created (hint: if you
haven't defined $TMPDIR, it's /tmp). Consider what filesystem that
directory is running.
2. Determine what the size of the file is that's being created
(compile the application with -g, run with gdb, determine the size,
etc).
3. Try creating the file with that size.

    Does step 3. pass or fail?
    Once you have these bits, please provide us with the details as
well as your kernel version, architecture, etc.
    We can give you the answer or teach you how to fish. I think that
teaching you how to fish is more worthwhile than fishing for you,
especially when LTP is a group of volunteers interested in testing on
Linux, not a support group.
    We can help you solve your problem if you get stuck, but please do
the initial legwork to help us help you.
Thanks,
-Garrett

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list