Lasse Kliemann | 3 Mar 2008 18:40

segfaults with gcc 4.2.3

Just for everyone's information: compiling star on newer gcc (I tested 
version 4.2.3 on Linux) with any optimization flags other than -O0 may 
result in segfaults in dump mode, like this:

   kern.info: star[4067]: segfault at 087f3c74 eip 0807b104 esp bf9878ac error 4

as well as this error:

   star: Success. Error moving extended header '././ <at> PaxHeader'.

in restore mode.

Compiling with

   smake COPTX=-O0

gives a working executable, however.

Maybe something can be done about it from star's side; otherwise everyone is 
advised not to use optimization with this gcc version.

Lasse
_______________________________________________
Star-users mailing list
Star-users <at> lists.berlios.de
https://lists.berlios.de/mailman/listinfo/star-users
Sven.Hartrumpf | 10 Mar 2008 14:34
Picon
Favicon

Re: segfaults with gcc 4.2.3

Mon, 3 Mar 2008 18:40:09 +0100, lasse-list-star-users wrote:

> Just for everyone's information: compiling star on newer gcc (I tested
> version 4.2.3 on Linux) with any optimization flags other than -O0 may
> result in segfaults in dump mode, like this:

How about gcc 4.3.0, which was released yesterday?
_______________________________________________
Star-users mailing list
Star-users <at> lists.berlios.de
https://lists.berlios.de/mailman/listinfo/star-users
Lasse Kliemann | 10 Mar 2008 20:13

Re: segfaults with gcc 4.2.3

* Message by -Sven.Hartrumpf <at> FernUni-Hagen.de- from Mon 2008-03-10:
> Mon, 3 Mar 2008 18:40:09 +0100, lasse-list-star-users wrote:
> 
> > Just for everyone's information: compiling star on newer gcc (I tested
> > version 4.2.3 on Linux) with any optimization flags other than -O0 may
> > result in segfaults in dump mode, like this:
> 
> How about gcc 4.3.0, which was released yesterday?

gcc 4.3.0 is fine; tested with star1.5a87 and a88.

_______________________________________________
Star-users mailing list
Star-users <at> lists.berlios.de
https://lists.berlios.de/mailman/listinfo/star-users
Joerg Schilling | 11 Mar 2008 18:36
Picon
Favicon

Re: segfaults with gcc 4.2.3

Lasse Kliemann <lasse-list-star-users <at> mail.plastictree.net> wrote:

> * Message by -Sven.Hartrumpf <at> FernUni-Hagen.de- from Mon 2008-03-10:
> > Mon, 3 Mar 2008 18:40:09 +0100, lasse-list-star-users wrote:
> > 
> > > Just for everyone's information: compiling star on newer gcc (I tested
> > > version 4.2.3 on Linux) with any optimization flags other than -O0 may
> > > result in segfaults in dump mode, like this:
> > 
> > How about gcc 4.3.0, which was released yesterday?
>
> gcc 4.3.0 is fine; tested with star1.5a87 and a88.

Good to hear, then we may hoe that we just need to wait until the broken 
optimizers will go away ;-)

Jörg

--

-- 
 EMail:joerg <at> schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
       js <at> cs.tu-berlin.de                (uni)  
       schilling <at> fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
Joerg Schilling | 3 Mar 2008 19:02
Picon
Favicon

Re: segfaults with gcc 4.2.3

Lasse Kliemann <lasse-list-star-users <at> mail.plastictree.net> wrote:

> Just for everyone's information: compiling star on newer gcc (I tested 
> version 4.2.3 on Linux) with any optimization flags other than -O0 may 
> result in segfaults in dump mode, like this:
>
>    kern.info: star[4067]: segfault at 087f3c74 eip 0807b104 esp bf9878ac error 4
>
> as well as this error:
>
>    star: Success. Error moving extended header '././ <at> PaxHeader'.
>
> in restore mode.

Is the problem also present with 4.0.2?

Jörg

--

-- 
 EMail:joerg <at> schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
       js <at> cs.tu-berlin.de                (uni)  
       schilling <at> fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
Lasse Kliemann | 3 Mar 2008 21:12

Re: segfaults with gcc 4.2.3

* Message by -Joerg Schilling- from Mon 2008-03-03:
> Lasse Kliemann <lasse-list-star-users <at> mail.plastictree.net> wrote:
> 
> > Just for everyone's information: compiling star on newer gcc (I tested 
> > version 4.2.3 on Linux) with any optimization flags other than -O0 may 
> > result in segfaults in dump mode, like this:
> >
> >    kern.info: star[4067]: segfault at 087f3c74 eip 0807b104 esp bf9878ac error 4
> >
> > as well as this error:
> >
> >    star: Success. Error moving extended header '././ <at> PaxHeader'.
> >
> > in restore mode.
> 
> Is the problem also present with 4.0.2?

No; tested on the same machine.

Is 4.0.2 a kind of 'gold' version?
The latest in the 4.0.x series is 4.0.4.

Lasse
_______________________________________________
Star-users mailing list
Star-users <at> lists.berlios.de
https://lists.berlios.de/mailman/listinfo/star-users
(Continue reading)

Joerg Schilling | 3 Mar 2008 21:14
Picon
Favicon

Re: segfaults with gcc 4.2.3

Lasse Kliemann <lasse-list-star-users <at> mail.plastictree.net> wrote:

> > > as well as this error:
> > >
> > >    star: Success. Error moving extended header '././ <at> PaxHeader'.
> > >
> > > in restore mode.
> > 
> > Is the problem also present with 4.0.2?
>
> No; tested on the same machine.
>
> Is 4.0.2 a kind of 'gold' version?
> The latest in the 4.0.x series is 4.0.4.

It is available in binary from Blastwave

Jörg

--

-- 
 EMail:joerg <at> schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
       js <at> cs.tu-berlin.de                (uni)  
       schilling <at> fokus.fraunhofer.de     (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily

Gmane