Your message dated Wed, 28 Sep 2011 06:52:26 +0200
with message-id <20110928045225.ga30...@hall.aurel32.net>
and subject line Re: /usr/bin/gij-4.6: segfault under kfreebsd-amd64
has caused the Debian Bug report #637474,
regarding /usr/bin/gij-4.6: segfault under kfreebsd-amd64
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
637474: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=637474
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: gcj-4.6-jre-headless
Version: 4.6.1-2
Severity: normal
File: /usr/bin/gij-4.6
X-Debbugs-CC: debian-...@lists.debian.org
Hi,
gij-4.6 just segfault for any java code under kfreebsd-amd64 (whereas it works
fine under kfreebsd-i386). I tested all this under asdfasdf.debian.net
porterbox.
Here is some example :
Test.java
----------------------------
public class Test {
public static void main(String[] args) {
System.out.println("OK");
}
}
----------------------------
$> /usr/lib/jvm/java-gcj-4.6/bin/javac Test.java
$> /usr/lib/jvm/java-gcj-4.6/bin/java Test
Segmentation fault
$>
Here is ktrace output for last fopen before crash :
----------------------------
59454 gij-4.6 CALL open(0x648bc0,O_RDONLY|O_LARGEFILE,<unused>0)
59454 gij-4.6 NAMI "/var/lib/gcj-4.6/classmap.db"
59454 gij-4.6 RET open 5
59454 gij-4.6 CALL stat(0x7fffffffe1a0,0x7fffffffe110)
59454 gij-4.6 NAMI "/var/lib/gcj-4.6/classmap.db"
59454 gij-4.6 STRU invalid record
59454 gij-4.6 RET stat 0
59454 gij-4.6 CALL fstat(0x5,0x7fffffffe1e0)
59454 gij-4.6 STRU invalid record
59454 gij-4.6 RET fstat 0
59454 gij-4.6 CALL mmap(0,0x4000,PROT_READ,MAP_PRIVATE,0x5,0)
59454 gij-4.6 RET mmap 34366324736/0x800648000
59454 gij-4.6 PSIG SIGSEGV SIG_DFL
----------------------------
So I thought it was linked to some previous mmap issues we already had (see
[1] or [2]) so I tried to skip mmap'ing of this classmap.db file (used for AOT-
Compile).
Sadly, it's not better :
$> /usr/lib/jvm/java-gcj-4.6/bin/java -Dgnu.gcj.precompiled.db.path= Test
Segmentation fault
$>
Ktrace :
----------------------------
75773 gij-4.6 NAMI "/home/drazzib/Test.class"
75773 gij-4.6 STRU invalid record
75773 gij-4.6 RET stat 0
75773 gij-4.6 CALL read(0x5,0x694c0c,0x194)
75773 gij-4.6 GIO fd 5 read 404 bytes
[...]
75773 gij-4.6 RET read 404/0x194
75773 gij-4.6 CALL close(0x5)
75773 gij-4.6 RET close 0
75773 gij-4.6 PSIG SIGSEGV SIG_DFL
----------------------------
Did someone have any clue ?
[1] http://bugs.debian.org/570889
[2] http://bugs.debian.org/571532
Regards,
--
Damien - Debian Developper
http://wiki.debian.org/DamienRaudeMorvan
signature.asc
Description: This is a digitally signed message part.
--- End Message ---
--- Begin Message ---
Version: 4.6.1-9
On Thu, Aug 11, 2011 at 11:11:50PM +0200, Damien Raude-Morvan wrote:
> Package: gcj-4.6-jre-headless
> Version: 4.6.1-2
> Severity: normal
> File: /usr/bin/gij-4.6
> X-Debbugs-CC: debian-...@lists.debian.org
>
> Hi,
>
> gij-4.6 just segfault for any java code under kfreebsd-amd64 (whereas it
> works
> fine under kfreebsd-i386). I tested all this under asdfasdf.debian.net
> porterbox.
>
> Here is some example :
> Test.java
> ----------------------------
> public class Test {
> public static void main(String[] args) {
> System.out.println("OK");
> }
> }
> ----------------------------
> $> /usr/lib/jvm/java-gcj-4.6/bin/javac Test.java
> $> /usr/lib/jvm/java-gcj-4.6/bin/java Test
> Segmentation fault
> $>
>
This has been fixed in version 4.6.1-9. Closing the bug.
--
Aurelien Jarno GPG: 1024D/F1BCDB73
aurel...@aurel32.net http://www.aurel32.net
--- End Message ---