[LTP] readahead02 failure on PPC

[Kshitij Malik]

Thank you Julio for the update.

Jan,

Can we take the changes shared by you as an official fix?

Thanks
KK
________________________________________
From: Julio Cruz Barroso <julio.cruz@smartmatic.com>
Sent: Wednesday, March 23, 2016 01:46
To: Kshitij Malik
Cc: Jan Stancek; ltp@lists.linux.it
Subject: RE: [LTP] readahead02 failure on PPC

Hi again,

Additionally, to previous comments, I try the patch in a 32bits system (with Kernel 3.14) and the TWARN is gone.

I perform the test 10 times in two boards: 1) 32 bits with one core and 512MB RAM and 2) 32 bits with two cores and same RAM:

- With patch: 5 times show PASS in both boards
- Without patch: 5 times show FAIL in both boards

This time, I don’t get inconsistent results.

Regards

Julio

> -----Original Message-----
> From: Kshitij Malik [mailto:Kshitij.Malik@mitel.com]
> Sent: Wednesday, March 23, 2016 10:55 AM
> To: Julio Cruz Barroso
> Cc: Jan Stancek; ltp@lists.linux.it
> Subject: RE: [LTP] readahead02 failure on PPC
>
> Hi Julio,
>
> Thank you for the update.
>
> I was also getting the /proc/pid/io error to start with and I fixed it by
> enabling kernel configurations CONFIG_TASKSTATS,
> CONFIG_TASK_IO_ACCOUNTING and other related configs.
>
> These configs creates the io directory for a thread in proc.
>
> Jan,
>
> Do you think your fix can be used as an official patch?
>
> Thanks
> KK
>
> On Mar 22, 2016 22:39, Julio Cruz Barroso <julio.cruz@smartmatic.com>
> wrote:
> >
> > Hi Jan, Kshitij,
> >
> > Checking my logs (imx6 32bits based platform) the error is similar [1]:
> >
> > readahead02 0 TWARN : readahead02.c:324: using less cache than
> expected
> >
> > However, the results were not consistent, because, once time PASS and
> another FAIL (with previous TWARN).
> >
> > Another thing different was (in my case):
> >
> > readahead02 3 TCONF : readahead02.c:309: Your system doesn't have
> /proc/pid/io, unable to determine read bytes during test
> >
> > This last one, is the part 1 of the test case (read bytes during test). The part
> 2 (verify the cache) is the same TWARN.
> >
> > At the end, I skipped this test case because the results were not consistent
> and need further analysis.
> >
> > Regards,
> > Julio
> >
> >
> > [1] logs for imx6: https://justpaste.it/s4zk
> >
> > -----Original Message-----
> > From: ltp [mailto:ltp-bounces+julio.cruz=smartmatic.com@lists.linux.it] On
> Behalf Of Julio Cruz Barroso
> > Sent: Tuesday, March 22, 2016 11:19 PM
> > To: Jan Stancek
> > Cc: Julio Cruz; Kshitij Malik; ltp@lists.linux.it
> > Subject: Re: [LTP] readahead02 failure on PPC
> >
> > Hi Jan, Kshitij,
> >
> > I just saw this email!
> >
> > I will review the issue because I cannot remember the result details and let
> you know tomorrow (CST).
> >
> > I expect also to upload the others patchs reviewed.
> >
> > Regards
> >
> > Julio
> >
> >
> > > On Mar 22, 2016, at 5:50 PM, Jan Stancek <jstancek@redhat.com> wrote:
> > >
> > > Kshitij
> >
> > --
> > Mailing list info: https://lists.linux.it/listinfo/ltp
> ________________________________
> NOTE: This e-mail (including any attachments) is for the sole use of the
> intended recipient(s) and may contain information that is confidential and/or
> protected by legal privilege. Any unauthorized review, use, copy, disclosure
> or distribution of this e-mail is strictly prohibited. If you are not the intended
> recipient, please notify Mitel immediately and destroy all copies of this e-
> mail. Mitel does not accept any liability for breach of security, error or virus
> that may result from the transmission of this message.