kernel-2.6.3-1.100 reiserfs ok, shmat strangeness (invalid arg)

[Bernhard K. Weisshuhn]

Hi *,

The sleep_on() dependency of reiserfs seems to be gone now, thanks Dave!
But something seems to have changed wrt shm: After booting the prebuild
smp kernel all applications failed on shmat(). Logs say something
like this:

  shmat(id=229378) failed: Invalid argument

After I compiled my own kernel from the same sources
(kernel-source-2.6.3-1.100), the problem went away. Everything runs
smooth now. Strange, huh?

Anyone want my .config?

cheers,
Bernie

[Dave Jones]

On Tue, 2004-02-24 at 19:45, Bernhard Weisshuhn wrote:
>  Hi *,
>  
>  The sleep_on() dependency of reiserfs seems to be gone now, thanks Dave! 
good, thanks.

>  But something seems to have changed wrt shm: After booting the
> prebuild
>  smp kernel all applications failed on shmat(). Logs say something
>  like this:
>  
>    shmat(id=229378) failed: Invalid argument 
looking into it. theres at least 1 report in bugzilla of this breaking X
too.

	Dave

[Steve Bergman]

Dave Jones wrote:

> >
> >  shmat(id=229378) failed: Invalid argument
> >    
> >
>
> looking into it. theres at least 1 report in bugzilla of this breaking X
> too.
>
>
>
>   
>  
X works for me.  But mplayer is broken:

X11 error: BadAccess (attempt to access private resource denied)

 From an strace:

2578  shmget(IPC_PRIVATE, 230400, IPC_CREAT|0777) = 1081356
2578  shmat(1081356, 0, 0)              = -1 EINVAL (Invalid argument)

My rawhide is coompletely up to date, except for openssl-devel which has 
a dependency issue.

-Steve

[Dave Jones]

On Wed, 2004-02-25 at 00:08, Steve Bergman wrote:

>  >
>  X works for me.  But mplayer is broken:
>  
>  X11 error: BadAccess (attempt to access private resource denied)
>  
>   From an strace:
>  
>  2578  shmget(IPC_PRIVATE, 230400, IPC_CREAT|0777) = 1081356
>  2578  shmat(1081356, 0, 0)              = -1 EINVAL (Invalid argument)
>  
>  My rawhide is coompletely up to date, except for openssl-devel which has 
>  a dependency issue. 
shmat was borked briefly. It should be fixed in rawhide now.

	Dave

davej(a)redhat.com (Dave Jones) writes:
>  looking into it. theres at least 1 report in bugzilla of this
> breaking X
>  too. 
Count me in...
-- 
Mathieu Chouquet-Stringer              E-Mail : mathieu(a)newview.com
       Never attribute to malice that which can be adequately
                    explained by stupidity.
                     -- Hanlon's Razor --

[Brian Masinick]

netopml(a)newview.com wrote:

> davej(a)redhat.com (Dave Jones) writes:
>   
>
> >looking into it. theres at least 1 report in bugzilla of this breaking X
> >too.
> >    
> >
>
> Count me in...
>   
>  
I have 2.6 kernels available for 3 systems; Fedora Core 2, Test 1, 
Mandrake 10.0 Beta 2, and Debian.  In all three cases, I can boot the 
system, but I can't run X.  Perhaps the problem is not unique to the 
Fedora Project.

-- 
Brian Masinick
mailto:[e-mail]

[Jim Cornette]

Brian Masinick wrote:
>  netopml(a)newview.com wrote:
>  
> > davej(a)redhat.com (Dave Jones) writes:
> >  
> >
> >> looking into it. theres at least 1 report in bugzilla of this breaking X
> >> too.
> >>   
> >
> >
> > Count me in...
> >  
> >
>  I have 2.6 kernels available for 3 systems; Fedora Core 2, Test 1, 
>  Mandrake 10.0 Beta 2, and Debian.  In all three cases, I can boot the 
>  system, but I can't run X.  Perhaps the problem is not unique to the 
>  Fedora Project.
>   

Reiserfs works now! But regarding dmesg relating to the X not starting 
problem. I am getting this excerpt in dmesg.

atkbd.c: Unknown key released (translated set 2, code 0x7a on 
isa0060/serio0).
atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly.
atkbd.c: Unknown key released (translated set 2, code 0x7a on 
isa0060/serio0).
atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly.


I'm running the smp version on a uniprocessor machine. I don't know if 
this happens with the non-smp kernel yet. X is working for me.

Maybe helpful, but probably nothing to do with X not starting on some 
machines.

Jim