[Bug runtime/17962] New: dtrace.exp regressed compared to rhel-6.6 package

[Bug runtime/17962] New: dtrace.exp regressed compared to rhel-6.6 package

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

            Bug ID: 17962
           Summary: dtrace.exp regressed compared to rhel-6.6 package
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: runtime
          Assignee: systemtap at sourceware dot org
          Reporter: mcermak at redhat dot com

On RHEL-6 the dtrace.exp testcase regressed (comparing latest upstream bits
against systemtap-2.5-5.el6 on some recent rhel-6.7 underlying OS).

The failure is "FAIL: dtrace parser check 1". On all RHEL-6 arches.

This is a rhel-6 regression. I can provide more details on demand, but I
believe the failure is easily reproducible.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp regressed compared to rhel-6.6 package

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

Martin Cermak <mcermak at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |scox at redhat dot com

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

Frank Ch. Eigler <fche at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fche at redhat dot com
            Summary|dtrace.exp regressed        |dtrace.exp --no-parsing
                   |compared to rhel-6.6        |fallback test fails on
                   |package                     |rhel6

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

--- Comment #1 from Frank Ch. Eigler <fche at redhat dot com> ---
This new subtest exercises the pyparsing -> no-pyparsing fallback
in dtrace.py.  This subtest fails on rhel6 because the pyparsing
package is not available in rhel6 (only epel6), so the fallback
cannot be triggered in the first place.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

Frank Ch. Eigler <fche at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #2 from Frank Ch. Eigler <fche at redhat dot com> ---
commit 4b0ab616d40a

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

Martin Cermak <mcermak at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |---

--- Comment #3 from Martin Cermak <mcermak at redhat dot com> ---
Looks much better now. However on s390x rhel6 and rhel7 I see following:

=======
 7.1 S s390x # cat ./repro.sh 
#!/bin/bash

# please adjust:

__PYTHON=/usr/bin/python2
__DTRACE=/root/mcermak-systemtap/systemtap-build/bin/dtrace

$__PYTHON $__DTRACE -C -G --no-pyparsing -s ./systemtap.base/dtrace.d -o
dtrace-1.o
$__PYTHON $__DTRACE -C -G --no-pyparsing -s ./systemtap.base/dtrace.d -o
dtrace-2.o

nm dtrace-1.o > dtrace-1.od
nm dtrace-2.o > dtrace-2.od

diff dtrace-1.od dtrace-2.od

 7.1 S s390x # ./repro.sh 
17c17
< 0000000000000176 N .LASF0
---
> 000000000000016f N .LASF0
19,20c19,20
< 0000000000000114 N .LASF10
< 0000000000000074 N .LASF11
---
> 00000000000000f4 N .LASF10
> 0000000000000156 N .LASF11
22,25c22,25
< 00000000000000a5 N .LASF13
< 0000000000000145 N .LASF14
< 00000000000000be N .LASF15
< 000000000000018e N .LASF16
---
> 0000000000000085 N .LASF13
> 0000000000000125 N .LASF14
> 000000000000009e N .LASF15
> 000000000000005c N .LASF16
27c27
< 0000000000000061 N .LASF18
---
> 0000000000000049 N .LASF18
31,34c31,34
< 000000000000015e N .LASF5
< 000000000000012d N .LASF6
< 0000000000000049 N .LASF7
< 000000000000008d N .LASF8
---
> 000000000000013e N .LASF5
> 0000000000000187 N .LASF6
> 000000000000010d N .LASF7
> 000000000000006d N .LASF8
 7.1 S s390x # 
=======

The diff is random among multiple runs. Quite seldom compared files are
identical. This reproducer only fails on rhel[67] s390x (not on rhel5 s390x).
It passes on all other rhel[567] platforms. On rhel7 I can install
pyparsing.rpm and remove `--no-pyparsing`, but this doesn't change the result.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[mcermak at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

--- Comment #4 from Martin Cermak <mcermak at redhat dot com> ---
Created attachment 8170
  --> https://sourceware.org/bugzilla/attachment.cgi?id=8170&action=edit
example dtrace output

Attached files created on el7 s390x thusly:

for i in `seq 0 9`; do python2 dtrace -C -G --no-pyparsing -s
./systemtap.base/dtrace.d -o dtrace-$i.o; done

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug runtime/17962] dtrace.exp --no-parsing fallback test fails on rhel6

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17962

Frank Ch. Eigler <fche at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #5 from Frank Ch. Eigler <fche at redhat dot com> ---
I don't see any problem with the varying output of the debugging-related
internal symbols of that object file.

-- 
You are receiving this mail because:
You are the assignee for the bug.