2003-08-20 21:31:17

by Paul Larson

Subject: LTP nightly regression results for 2.6.0-test3-bk2,bk3,bk5,bk6,bk7,mm3,mjb1

Here's the latest batch of LTP test results for the latest 2.6 bk
snapshots and mm/mjb trees. These results and previous results are
archived at: http://developer.osdl.org/dev/ltp/results/

Thanks,
Paul Larson

2.6.0-test3-bk1-vs-2.6.0-test3-bk2
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk1-vs-2.6.0-test3-bk2/
Test Name 2.6.0-test3-bk1 2.6.0-test3-bk2 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
nanosleep02 FAIL FAIL N N
setegid01 FAIL PASS N Y
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N

2.6.0-test3-bk2-vs-2.6.0-test3-bk3
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk2-vs-2.6.0-test3-bk3/
Test Name 2.6.0-test3-bk2 2.6.0-test3-bk3 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
gettimeofday PASS FAIL Y N
nanosleep02 FAIL FAIL N N
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N
syslog04 PASS FAIL Y N

*** bk4 failed to compile, but the failure was quickly fixed in bk5 ***

2.6.0-test3-bk3-vs-2.6.0-test3-bk5
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk3-vs-2.6.0-test3-bk5/
Test Name 2.6.0-test3-bk3 2.6.0-test3-bk5 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
gettimeofday FAIL PASS N Y
nanosleep02 FAIL FAIL N N
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N
syslog04 FAIL FAIL N N

2.6.0-test3-bk5-vs-2.6.0-test3-bk6
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk5-vs-2.6.0-test3-bk6/
Test Name 2.6.0-test3-bk5 2.6.0-test3-bk6 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
nanosleep02 FAIL FAIL N N
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N
syslog04 FAIL PASS N Y

2.6.0-test3-bk6-vs-2.6.0-test3-bk7
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk6-vs-2.6.0-test3-bk7/
Test Name 2.6.0-test3-bk6 2.6.0-test3-bk7 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
gettimeofday PASS FAIL Y N
nanosleep02 FAIL FAIL N N
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N
syslog04 PASS FAIL Y N

2.6.0-test3-mm2-vs-2.6.0-test3-mm3
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-mm2-vs-2.6.0-test3-mm3/
Test Name 2.6.0-test3-mm2 2.6.0-test3-mm3 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
nanosleep02 FAIL FAIL N N
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N

2.6.0-test3-vs-2.6.0-test3-mjb1
http://developer.osdl.org/dev/ltp/results/2.6.0-test3-vs-2.6.0-test3-mjb1/
Test Name 2.6.0-test3 2.6.0-test3-mjb1 Regression Improvement
-----------------------------------------------------------------------------
alarm03 FAIL FAIL N N
getgroups03 FAIL FAIL N N
nanosleep02 FAIL FAIL N N
setegid01 FAIL PASS N Y
swapoff01 FAIL FAIL N N
swapoff02 FAIL FAIL N N

2003-08-21 04:57:07

by George Anzinger

Subject: Re: LTP nightly regression results for 2.6.0-test3-bk2,bk3,bk5,bk6,bk7,mm3,mjb1

Paul Larson wrote:
>
> Here's the latest batch of LTP test results for the latest 2.6 bk
> snapshots and mm/mjb trees. These results and previous results are
> archived at: http://developer.osdl.org/dev/ltp/results/
>
> Thanks,
> Paul Larson

Ok, so suppose I want to tackle one (or more) of these. Where do I
find the sources for the test. And if I find the test in error, what
do I do?


>
> 2.6.0-test3-bk1-vs-2.6.0-test3-bk2
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk1-vs-2.6.0-test3-bk2/
> Test Name 2.6.0-test3-bk1 2.6.0-test3-bk2 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> nanosleep02 FAIL FAIL N N
> setegid01 FAIL PASS N Y
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N

--
George Anzinger [email protected]
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

2003-08-21 09:25:29

by George Anzinger

Subject: Re: LTP nightly regression results for 2.6.0-test3-bk2,bk3,bk5,bk6,bk7,mm3,mjb1

Paul Larson wrote:
>
> Here's the latest batch of LTP test results for the latest 2.6 bk
> snapshots and mm/mjb trees. These results and previous results are
> archived at: http://developer.osdl.org/dev/ltp/results/

The failure in alarm03, which by the way is only some times, is caused
by the need to add a jiffie to the alarm time to cover the fact that
the current time is not on a jiffie boundry. If this was not done,
the timer would expire early about 50% of the time. Given this
correction the timer will expire some where between the requested time
and the requested time + one jiffie. alarm(2), dealing with seconds,
is bumping the return value by one when ever the return is not an even
second, thus the 1 second longer than requested result.

In short, the test should allow the return value to be one higher than
requested.
-g
>
> Thanks,
> Paul Larson
>
> 2.6.0-test3-bk1-vs-2.6.0-test3-bk2
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk1-vs-2.6.0-test3-bk2/
> Test Name 2.6.0-test3-bk1 2.6.0-test3-bk2 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> nanosleep02 FAIL FAIL N N
> setegid01 FAIL PASS N Y
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
>
> 2.6.0-test3-bk2-vs-2.6.0-test3-bk3
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk2-vs-2.6.0-test3-bk3/
> Test Name 2.6.0-test3-bk2 2.6.0-test3-bk3 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> gettimeofday PASS FAIL Y N
> nanosleep02 FAIL FAIL N N
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
> syslog04 PASS FAIL Y N
>
> *** bk4 failed to compile, but the failure was quickly fixed in bk5 ***
>
> 2.6.0-test3-bk3-vs-2.6.0-test3-bk5
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk3-vs-2.6.0-test3-bk5/
> Test Name 2.6.0-test3-bk3 2.6.0-test3-bk5 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> gettimeofday FAIL PASS N Y
> nanosleep02 FAIL FAIL N N
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
> syslog04 FAIL FAIL N N
>
> 2.6.0-test3-bk5-vs-2.6.0-test3-bk6
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk5-vs-2.6.0-test3-bk6/
> Test Name 2.6.0-test3-bk5 2.6.0-test3-bk6 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> nanosleep02 FAIL FAIL N N
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
> syslog04 FAIL PASS N Y
>
> 2.6.0-test3-bk6-vs-2.6.0-test3-bk7
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-bk6-vs-2.6.0-test3-bk7/
> Test Name 2.6.0-test3-bk6 2.6.0-test3-bk7 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> gettimeofday PASS FAIL Y N
> nanosleep02 FAIL FAIL N N
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
> syslog04 PASS FAIL Y N
>
> 2.6.0-test3-mm2-vs-2.6.0-test3-mm3
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-mm2-vs-2.6.0-test3-mm3/
> Test Name 2.6.0-test3-mm2 2.6.0-test3-mm3 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> nanosleep02 FAIL FAIL N N
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
>
> 2.6.0-test3-vs-2.6.0-test3-mjb1
> http://developer.osdl.org/dev/ltp/results/2.6.0-test3-vs-2.6.0-test3-mjb1/
> Test Name 2.6.0-test3 2.6.0-test3-mjb1 Regression Improvement
> -----------------------------------------------------------------------------
> alarm03 FAIL FAIL N N
> getgroups03 FAIL FAIL N N
> nanosleep02 FAIL FAIL N N
> setegid01 FAIL PASS N Y
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>

--
George Anzinger [email protected]
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

2003-08-25 16:09:49

by Hugh Dickins

Subject: Re: LTP nightly regression results for 2.6.0-test3-bk2,bk3,bk5,bk6,bk7,mm3,mjb1

On 20 Aug 2003, Paul Larson wrote:
>
> Here's the latest batch of LTP test results for the latest 2.6 bk
> snapshots and mm/mjb trees. These results and previous results are
> archived at: http://developer.osdl.org/dev/ltp/results/
>....
> swapoff01 FAIL FAIL N N
> swapoff02 FAIL FAIL N N

Paul, I've not been able to reproduce those swapoff failures.
Your ltp.outs indicate that it was the swapon which failed,
but its error code is not shown; nor the tmpfile outputs from
setup's system(3) calls (if I remember rightly, system succeeds
if it can run what's asked, whether what it runs fails or not).
Please try running a swapoff01 and swapoff02 modified to catch
that info: it may quickly identify the issue.

Hugh