Test results on 2.6.28-rc8

Test results on 2.6.28-rc8

[Wenji Huang]

Hi,

Test results on 2.6.28-rc8.
Seems problem with backtrace/context.

Regards,
wenji

------------------------------------------------------------------------------
Package:        systemtap_snapshot 
(0.8/0.131-master(04844b33)-2.6.28_rc8-snapshot)

Previous build: 2008/12/01 23:34:19 - 2008/12/01 23:34:19
Current build:  2008/12/10 16:01:09 - 2008/12/10 16:01:09
Host:           rctest-32
Platform:       Linux 2.6.28-rc8 i386 (EL4U6) (before: Linux 2.6.28-rc7 
i386 (EL4U6))
URL: 
http://build.alchar.org/cgi/showBuild?host=rctest-32&pkg=systemtap_snapshot&date=20081210-160109
First failure:  test
Test result:    829: 757 +, 44 -, 28 S, 0 E (before: 816: 727 +, 61 -, 
28 S, 0 E)

   Old   -> New   Test
   -----    ----- ----------------------------------
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse13)
   FAIL  -> N/A   test - systemtap.base/marker.exp(K_MARKER18 was cached)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit uid16)
   FAIL  -> N/A   test - systemtap.base/marker.exp(K_MARKER11 was cached)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse8)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse4)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit sendfile)
   N/A   -> PASS  test - systemtap.base/marker.exp(K_MARKER04 wasn't cached)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit mount)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit trunc)
   N/A   -> PASS  test - systemtap.base/marker.exp(K_MARKER18 wasn't cached)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func4)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit unlink)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit signal)
   N/A   -> PASS  test - systemtap.base/backtrace.exp(backtrace-unwindsyms)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse14)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse10)
   FAIL  -> N/A   test - systemtap.base/marker.exp(K_MARKER04 was cached)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse9)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse5)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func2)
   FAIL  -> PASS  test - systemtap.base/stmtvars.exp(stmtvars - .function)
   PASS  -> FAIL  test - systemtap.base/alternatives.exp(LOCAL1)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit swap)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit statfs)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit select)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit stat)
   N/A   -> PASS  test - systemtap.base/marker.exp(K_MARKER11 wasn't cached)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit uid)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse11)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit rt_signal)
   N/A   -> PASS  test - systemtap.base/global_stat.exp(global_stat)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit openclose)
   PASS  -> FAIL  test - systemtap.base/debugpath.exp(debugpath-good)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse6)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse2)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func3)
   PASS  -> N/A   test - systemtap.context/context.exp(backtrace of 
begin probe)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit poll)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit umask)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func2)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit mmap)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse12)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit sync)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse7)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse3)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func4)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit readwrite)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit timer)
   PASS  -> FAIL  test - systemtap.syscall/syscall.exp(32-bit acct)
   N/A   -> PASS  test - systemtap.base/backtrace.exp(backtrace)
   FAIL  -> PASS  test - systemtap.base/stmtvars.exp(stmtvars)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func3)
   FAIL  -> N/A   test - systemtap.base/cmd_parse.exp(cmd_parse2: 
unexpected EOF)

------------------------------------------------------------------------------
Package:        systemtap_snapshot 
(0.8/0.131-master(c8c12f3c)-2.6.28_rc8-snapshot)

Previous build: 2008/12/02 15:26:45 - 2008/12/02 15:26:45
Current build:  2008/12/11 12:56:17 - 2008/12/11 12:56:17
Host:           rctest-64
Platform:       Linux 2.6.28-rc8 x86_64 (EL5.2) (before: Linux 
2.6.28-rc7 x86_64 (EL5.2))
URL: 
http://build.alchar.org/cgi/showBuild?host=rctest-64&pkg=systemtap_snapshot&date=20081211-125617
First failure:  test
Test result:    864: 790 +, 44 -, 30 S, 0 E (before: 849: 775 +, 42 -, 
32 S, 0 E)

   Old   -> New   Test
   -----    ----- ----------------------------------
   N/A   -> FAIL  test - systemtap.syscall/syscall.exp(64-bit itimer)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse13)
   FAIL  -> N/A   test - systemtap.base/cmd_parse.exp(cmd_parse1: 
unexpected EOF)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse8)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse4)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit sendfile)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit trunc)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func4)
   N/A   -> FAIL  test - systemtap.base/backtrace.exp(backtrace-unwindsyms)
   SKIP  -> N/A   test - systemtap.base/marker.exp(K_MARKER09 : no 
kernel marker found with a numeric first argument)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse14)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse10)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse9)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse5)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse1)
   FAIL  -> N/A   test - systemtap.syscall/syscall.exp(64-bit signal)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func2)
   N/A   -> FAIL  test - 
systemtap.stress/current.exp(systemtap.stress/current.stp shutdown)
   PASS  -> N/A   test - 
systemtap.stress/current.exp(systemtap.stress/current.stp shutdown and 
output)
   PASS  -> N/A   test - systemtap.syscall/syscall.exp(itimer)
   FAIL  -> PASS  test - systemtap.base/warnings.exp(warnings)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit uid)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse11)
   N/A   -> PASS  test - systemtap.base/global_stat.exp(global_stat)
   SKIP  -> N/A   test - systemtap.base/marker.exp(K_MARKER10 : no 
kernel marker found with a numeric first argument)
   PASS  -> FAIL  test - systemtap.base/debugpath.exp(debugpath-good)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse6)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse2)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func3)
   PASS  -> N/A   test - systemtap.context/context.exp(backtrace of 
begin probe)
   N/A   -> PASS  test - systemtap.base/marker.exp(K_MARKER09 wasn't cached)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func2)
   FAIL  -> N/A   test - systemtap.syscall/syscall.exp(64-bit umask)
   FAIL  -> PASS  test - systemtap.base/beginenderror.exp(beginenderror)
   N/A   -> PASS  test - systemtap.base/marker.exp(K_MARKER10 wasn't cached)
   N/A   -> PASS  test - systemtap.syscall/syscall.exp(signal)
   N/A   -> PASS  test - systemtap.syscall/syscall.exp(rt_signal)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse12)
   FAIL  -> PASS  test - systemtap.syscall/syscall.exp(32-bit sync)
   N/A   -> PASS  test - systemtap.syscall/syscall.exp(umask)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse7)
   N/A   -> PASS  test - systemtap.base/cmd_parse.exp(cmd_parse3)
   PASS  -> FAIL  test - systemtap.syscall/syscall.exp(32-bit chmod)
   PASS  -> FAIL  test - systemtap.context/context.exp(print_stack of 
yyy_func4)
   N/A   -> FAIL  test - systemtap.base/backtrace.exp(backtrace)
   PASS  -> FAIL  test - systemtap.context/context.exp(backtrace of 
yyy_func3)
   FAIL  -> N/A   test - systemtap.syscall/syscall.exp(64-bit rt_signal)

Re: Test results on 2.6.28-rc8

[Mark Wielaard]

Hi Wenji,

On Thu, 2008-12-18 at 11:29 +0800, Wenji Huang wrote:
> Test results on 2.6.28-rc8.
> Seems problem with backtrace/context.
>
> ------------------------------------------------------------------------------
> Package:        systemtap_snapshot 
> (0.8/0.131-master(04844b33)-2.6.28_rc8-snapshot)
> 
> Previous build: 2008/12/01 23:34:19 - 2008/12/01 23:34:19
> Current build:  2008/12/10 16:01:09 - 2008/12/10 16:01:09
> Host:           rctest-32
> Platform:       Linux 2.6.28-rc8 i386 (EL4U6) (before: Linux 2.6.28-rc7 
> i386 (EL4U6))
> URL: 
> http://build.alchar.org/cgi/showBuild?host=rctest-32&pkg=systemtap_snapshot&date=20081210-160109
> First failure:  test
> Test result:    829: 757 +, 44 -, 28 S, 0 E (before: 816: 727 +, 61 -, 
> 28 S, 0 E)

The above URL doesn't seem to contain the full testsuite
systemtap.log/.sum files. Could you post them, so we can inspect the
specific failures?

Thanks,

Mark

Re: Test results on 2.6.28-rc8

[Kris Van Hees]

Mark,

The .sum file is now available for every systemtap_snapshot reported on labrat.
Just look for the 'Download build log' link.  Note that it only contains the
.sum file information.  I have found the .log very useful in the past for
tracking down test failures, but past discussion within the systemtap group
resulted in objections against using the .log file.  If you prefer to have
the .log available, we can definitely start doing that going forward (as we
used to do that in the past anyway).

	Cheers,
	Kris

On Thu, Dec 18, 2008 at 03:49:16PM +0100, Mark Wielaard wrote:
> Hi Wenji,
> 
> On Thu, 2008-12-18 at 11:29 +0800, Wenji Huang wrote:
> > Test results on 2.6.28-rc8.
> > Seems problem with backtrace/context.
> >
> > ------------------------------------------------------------------------------
> > Package:        systemtap_snapshot 
> > (0.8/0.131-master(04844b33)-2.6.28_rc8-snapshot)
> > 
> > Previous build: 2008/12/01 23:34:19 - 2008/12/01 23:34:19
> > Current build:  2008/12/10 16:01:09 - 2008/12/10 16:01:09
> > Host:           rctest-32
> > Platform:       Linux 2.6.28-rc8 i386 (EL4U6) (before: Linux 2.6.28-rc7 
> > i386 (EL4U6))
> > URL: 
> > http://build.alchar.org/cgi/showBuild?host=rctest-32&pkg=systemtap_snapshot&date=20081210-160109
> > First failure:  test
> > Test result:    829: 757 +, 44 -, 28 S, 0 E (before: 816: 727 +, 61 -, 
> > 28 S, 0 E)
> 
> The above URL doesn't seem to contain the full testsuite
> systemtap.log/.sum files. Could you post them, so we can inspect the
> specific failures?
> 
> Thanks,
> 
> Mark
> 

Re: Test results on 2.6.28-rc8

[Mark Wielaard]

Hi Kris,

On Thu, 2008-12-18 at 10:51 -0500, Kris Van Hees wrote:
> The .sum file is now available for every systemtap_snapshot reported on labrat.
> Just look for the 'Download build log' link.  Note that it only contains the
> .sum file information.  I have found the .log very useful in the past for
> tracking down test failures, but past discussion within the systemtap group
> resulted in objections against using the .log file.  If you prefer to have
> the .log available, we can definitely start doing that going forward (as we
> used to do that in the past anyway).

Thanks! But indeed the testsuite/systemtap.log file would be more
useful. So if you could include that, that would be highly appreciated.

Cheers,

Mark

Re: Test results on 2.6.28-rc8

[Frank Ch. Eigler]

Kris Van Hees <kris.van.hees@oracle.com> writes:

> [...]  I have found the .log very useful in the past for tracking
> down test failures, but past discussion within the systemtap group
> resulted in objections against using the .log file.  [...]

I recall the issue was whether emailing the log file to labrat was
appropriate, considering that it can be obnoxiously large, and
mechanically parsing that is less reliable than the .sum file.  If for
your testing, the file is economical to collect and store for
additional information, I can't think of a reason not to use it for
reference.

- FChE