Opened 9 years ago

Closed 7 years ago

Last modified 6 years ago

#406 closed bug (worksforme)

internal error: EVACUATED object entered!

Reported by: malaire Owned by: simonmar
Priority: normal Milestone: 6.6.1
Component: None Version: 6.4
Keywords: Cc:
Operating System: Unknown/Multiple Architecture: Unknown/Multiple
Type of failure: Difficulty: Unknown
Test Case: Blocked By:
Blocking: Related Tickets:

Description (last modified by igloo)

During 'make test' of pugs (http://pugscode.org/) which
is written in Haskell I got this error:

...
...
t/oo/construction..................................ok
        6/7 TODO feature tests
t/oo/delegation....................................ok
        24/34 TODO feature tests
t/oo/destruction...................................ok
6/6# Looks like you failed 2 tests of 6
pugs: internal error: EVACUATED object entered!
    Please report this as a bug to
glasgow-haskell-bugs@haskell.org,
    or http://www.sourceforge.net/projects/ghc/
t/oo/destruction...................................dubious
        Test returned status 254 (wstat 65024, 0xfe00)
DIED. FAILED tests 4-5
        Failed 2/6 tests, 66.67% okay
t/oo/enums.........................................ok
        54/58 TODO feature tests
...
...


I have:
* GHC 6.4
* pugs r4916
* (no parrot)
* SuSe 9.1
* "uname -a" gives "Linux markus 2.6.5-7.155.29-default
#1 Thu Jun 2 12:07:05 UTC 2005 i686 athlon i386 GNU/Linux"

Change History (10)

comment:1 Changed 9 years ago by malaire

Logged In: YES 
user_id=331912

This just might be related to bug [ 1222522 ] which I got at
the same test-file with earlier version of pugs.

comment:2 Changed 9 years ago by simonmar

Logged In: YES 
user_id=48280

As a data point, I don't get this error with:
  
  Pugs 6.2.7
  GHC 6.4.1-20050705 on x86_64-unknown-linux

That doesn't necessarily mean it's fixed, though.  It might
be a sensitive GC bug, or it could be a bug in Pugs.  I'll
see if I can get closer to your setup to reproduce it.

comment:3 Changed 9 years ago by malaire

Logged In: YES 
user_id=331912

I just did a double-check with

> svn co http://svn.openfoundry.org/pugs -r 4916
> cd pugs
> perl Makefile.PL && make
> make test

and got the same error.

I've updated pugs quite often and havn't got this error with
other revisions.

comment:4 Changed 8 years ago by simonmar

Logged In: YES 
user_id=48280

I'm reasonably sure I've fixed this, in rev. 1.276 of
ghc/rts/Schedule.c.  I wasn't able to reproduce the symptoms
using Pugs, but it is highly likely that this is the same
bug - triggered duing a GC near program termination when
there are finalizers running.

comment:5 Changed 8 years ago by simonmar

  • Status changed from assigned to closed

comment:6 Changed 8 years ago by markstos

  • Architecture set to Unknown
  • Difficulty set to Unknown
  • Operating System set to Unknown
  • Resolution Fixed deleted
  • Status changed from closed to reopened

I got this error today with GHC 6.4.2 and Pugs 6.2.12. Here's the script I ran to trigger it:

use v6-alpha;

class mark_test

has $!foo is rw;

method first {

say $!foo;

}

method second {

$!foo = 'zoo';

}

mark_test.new.second;

comment:7 Changed 7 years ago by igloo

  • Description modified (diff)
  • Milestone set to 6.6.1

Does anyone know if this is fixed in 6.6?

comment:8 Changed 7 years ago by simonmar

  • Resolution set to worksforme
  • Status changed from reopened to closed

Tried to reproduce without success using GHC 6.6. I downloaded the exact revision of Pugs (4916), and fixed up a few things in the source to make it compile with 6.6. The oo/destruction tests appear to pass without failure.

comment:9 Changed 6 years ago by simonmar

  • Architecture changed from Unknown to Unknown/Multiple

comment:10 Changed 6 years ago by simonmar

  • Operating System changed from Unknown to Unknown/Multiple
Note: See TracTickets for help on using tickets.