The kernel locks up completely whenever I launch any particularly large
application under X (xterm is fine, netscape locks up the box).
I've confirmed that this isn't just X locking up, as the machine is
completely frozen (doesn't respond to pings, doesn't respond to three
finger salute, etc).
Just to make sure that this isn't a VM/Memory problem:
I've run (and the machine passed) memtest86.
I've tried running X with swapoff (and the machine still locks up).
- Kernel 2.4.19
- GLIBC 2.2.90
- XFree86 4.2.0
I can reproduce this error every single time I run X. However, I have
not been able to see the problem on the shell.
Anyone ever see this problem? Does anyone have pointers on how I can
better troubleshoot the problem?
On Wed, 2002-09-04 at 02:36, John Weber wrote:
> The kernel locks up completely whenever I launch any particularly large
> application under X (xterm is fine, netscape locks up the box).
> I've confirmed that this isn't just X locking up, as the machine is
> completely frozen (doesn't respond to pings, doesn't respond to three
> finger salute, etc).
Can you duplicate this with the vesa and/or vga16 drivers ?
Alan Cox wrote:
> On Wed, 2002-09-04 at 02:36, John Weber wrote:
>
>>The kernel locks up completely whenever I launch any particularly large
>>application under X (xterm is fine, netscape locks up the box).
>>I've confirmed that this isn't just X locking up, as the machine is
>>completely frozen (doesn't respond to pings, doesn't respond to three
>>finger salute, etc).
>
>
> Can you duplicate this with the vesa and/or vga16 drivers ?
>
The kernel locks up regardless of what drivers I use (and I have tried
the vga16 drivers already).
Disabling the accelerated functions also "fixes" the machine, for some
definitions of "fix" :).
My question is really why this problem would lock up the kernel...
I can't really tell whether the problem is in the implementation of the
XFree86 functions or the kernel functions it is calling, but the fact
that the entire kernel locks up suggests that both are to blame. Can
you suggest where I should start reading code (if not the file atleast
the directory :).
(o- j o h n e w e b e r
//\ aspiring computer scientist
v_/_ http://www-cs.ccny.cuny.edu/acm/weber/
> Disabling the accelerated functions also "fixes" the machine, for some
> definitions of "fix" :).
That sounds like an X bug then
> My question is really why this problem would lock up the kernel...
> I can't really tell whether the problem is in the implementation of the
> XFree86 functions or the kernel functions it is calling, but the fact
> that the entire kernel locks up suggests that both are to blame. Can
> you suggest where I should start reading code (if not the file atleast
> the directory :).
X maps the video hardware and drives it directly. In that sense X is the
device driver for video not the kernel.
On Thu, Sep 05, 2002 at 04:49:35PM +0100, Alan Cox wrote:
> > Disabling the accelerated functions also "fixes" the machine, for some
> > definitions of "fix" :).
>
> That sounds like an X bug then
>
> > My question is really why this problem would lock up the kernel...
> > I can't really tell whether the problem is in the implementation of the
> > XFree86 functions or the kernel functions it is calling, but the fact
> > that the entire kernel locks up suggests that both are to blame. Can
> > you suggest where I should start reading code (if not the file atleast
> > the directory :).
>
> X maps the video hardware and drives it directly. In that sense X is the
> device driver for video not the kernel.
... And some chips have bugs like "will hang the PCI bus if you
disable this while the chip is doing that". The chip "errata" will
read "don't do that then", and the manufacturer will try to make the
windows driver "not do that". But the Xfree driver may be lagging and
still do this occasionally.
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* The Worlds Ecosystem is a stable system. Stable systems may experience *
* excursions from the stable situation. We are currenly in such an *
* excursion: The stable situation does not include humans. ***************