bugPSPP - Bugs: bug #58943, Test 127 "zero or one...

 
 

bug #58943: Test 127 "zero or one variable in mrset" fails on debian/bullseye with -O2 -g optimization

Submitter:  Friedrich Beckmann <beckmanf>
Submitted:  Thu 13 Aug 2020 12:54:19 PM UTC
   
 
Category:  Compilation/Portability Severity:  5 - Average
Status:  Fixed Assigned to:  None
Open/Closed:  Closed Release:  None
Effort:  0.00
* Mandatory Fields

Add a New Comment Rich Markup
   

Fri 14 Aug 2020 01:07:10 AM UTC, comment #2: 

I fixed this with

https://git.savannah.gnu.org/cgit/pspp.git/commit/?id=8c76dd167f2b63b3c97606c53667c391c17fae10

We could maybe initialize all loop local variables to see this kind of problem easier.

Friedrich Beckmann <beckmanf>
Group Member
Thu 13 Aug 2020 02:51:35 PM UTC, comment #1: 

I added debian-sid to the continuous integration. See the failed regression:

http://caeis.etech.fh-augsburg.de:8010/#/builders/3/builds/2

Friedrich Beckmann <beckmanf>
Group Member
Thu 13 Aug 2020 12:54:19 PM UTC, original submission:  

The current pspp git master on commit e45cdf7b0e1b1b fails on debian/bullseye in test 127 when pspp is compiled with opimization "-O2 -g".

The failed test "testing zero or one variable in mrset " produces the following output:


#                             -*- compilation -*-
127. sys-file-reader.at:2780: testing zero or one variable in mrset ...
../../pspp/tests/sys-file-reader.at:2799: sack --$variant sys-file.sack > sys-file.sav
../../pspp/tests/sys-file-reader.at:2802: pspp -O format=csv sys-file.sps
--- -   2020-08-13 14:17:37.563145916 +0200
+++ /home/fritz/pspp/build/tests/testsuite.dir/at-groups/127/stdout     2020-08-13 14:17:37.555867407 +0200
@@ -1,3 +1,5 @@
+warning: `sys-file.sav' near offset 0xd8: Missing new-line parsing variable names at offset 22 in MRSETS record.
+
 warning: `sys-file.sav': MRSET $a has only one variable.

 warning: `sys-file.sav': MRSET $b has no variables.
127. sys-file-reader.at:2780: 127. zero or one variable in mrset (sys-file-reader.at:2780): FAILED (sys-file-reader.at:2802)


The test succeeds when the compiler setting is changed to "-O0 -g". The same happens on the 1.2.0 release. There were reports about the same failed test for 1.2.0 on ppc64el architecture when compiled with "-O3" from Steve Langasek:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=942609

That was fixed by changing the optimization to "-O2". Now we have this problem on amd64 when compiler optimization is "-O2". Either it is a compiler bug or we have some unconventional code...

The sack seems to be o.k. because I can read the produced sys-file.sav file with a "-O0 -g" pspp version and the result is o.k.

The gcc compiler version on debian/bullseye is: gcc (Debian 10.1.0-6) 10.1.0 but the problem report on ppc64el was with gcc-8.

Friedrich Beckmann <beckmanf>
Group Member

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by beckmanf (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only logged-in users can vote.

     

    Follow 2 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2020-08-14 beckmanf StatusNone Fixed
        Open/ClosedOpen Closed

    Back to the top

    Powered by Savane 3.13-f8d8.
    Corresponding source code