On 7/13/06, Jeff Dike <[EMAIL PROTECTED]> wrote:
On Mon, Jul 10, 2006 at 02:38:07AM -0400, Samuel Korpi wrote:
> > It's a bug, regardless. Can you attach gdb to it and get a stacktrace?
> >
>
> Stacktrace shows this:
>
> (gdb) bt
> #0 0x60140149 in __const_udelay (usecs=26734100480)
HiI have been trying to debug guest uml (2.6.16 patched with uml-2.6.16-bs2.patch) by setting breakpoints at start_kernel() but while running under host kernel (2.6.16.-skas3-v9-pre9) , it ignores the breakpoints and boots to login prompt.On the other hand, when i ran the debug session on my FC5 d
On Sun, Jul 09, 2006 at 07:16:58PM +0530, Ritesh Raj Sarraf wrote:
> On my guest kernel, I periodically see soft lockups. They don't
> crash the guest OS. These lockups, I believe, happen only when under
> heady disk I/O load.
They are annoying, but harmless. I haven't figured out what is going
o
On Wed, Jul 12, 2006 at 09:05:22AM +0800, Miao Qingjun wrote:
> I start "gdb uml", "b sys_init_module", then "run",
> but gdb is not break at sys_init_module() when I insmod LKM?
> Why? How can I do it?
Some versions of gdb don't work well on UML. One thing that often
works is to attach to UML wh
On Wed, Jul 12, 2006 at 12:21:42PM +0300, Hayim Shaul wrote:
> I copied the defines:
>
> #if defined __USE_MISC || defined _ASM
> # define JB_BX 0
> # define JB_SI 1
> # define JB_DI 2
> # define JB_BP 3
> # define JB_SP 4
> # define JB_PC 5
> # define JB_SIZE 24
> #endif
Instead of the pat
On Wed, Jul 12, 2006 at 05:14:17PM +0200, Josef Spillner wrote:
> The uml_net helper tool was indeed in /usr/lib/uml, and moving it fixed the
> hang. Isn't there a better error indiciation than just hanging in the case of
> the utility not being present?
Yes, there is. This was a stupid bug, wil
On Thu, Jul 13, 2006 at 08:50:29PM +0200, Blaisorblade wrote:
> SIGTRAP must be sent for the breakpoint - I think the problem is that either
> GDB misinterprets it or it is not sent the correct way.
I'm thinking the latter...
Jeff
---
On Thursday 13 July 2006 20:45, Jeff Dike wrote:
> On Thu, Jul 13, 2006 at 08:49:31PM +0800, Miao Qingjun wrote:
> > Under UML bash, I gdb a multithreading program, and set a breakpoint on
> > thread entry function, But when run it, gdb is not stop at the
> > breakpoint,
> > instead, program exit b
On Thu, Jul 13, 2006 at 08:49:31PM +0800, Miao Qingjun wrote:
> Under UML bash, I gdb a multithreading program, and set a breakpoint on
> thread entry function,
> But when run it, gdb is not stop at the breakpoint,
> instead, program exit because:
>
> Program terminated with signal SIGTRAP, Trac
On Thursday 13 July 2006 14:49, Miao Qingjun wrote:
> Hi,
>
> Under UML bash, I gdb a multithreading program, and set a breakpoint on
> thread entry function, But when run it, gdb is not stop at the breakpoint,
> instead, program exit because:
>
> Program terminated with signal SIGTRAP, Trace/break
On Mon, Jul 10, 2006 at 02:38:07AM -0400, Samuel Korpi wrote:
> > It's a bug, regardless. Can you attach gdb to it and get a stacktrace?
> >
>
> Stacktrace shows this:
>
> (gdb) bt
> #0 0x60140149 in __const_udelay (usecs=26734100480)
> at arch/um/sys-x86_64/delay.c:37
> #1 0x0
Hi,
Under UML bash, I gdb a multithreading program, and set a
breakpoint on thread entry function,
But when run it, gdb is not stop at the breakpoint,
instead, program exit because:
Program terminated with signal SIGTRAP, Trace/breakpoint
trap.
Why?
Thanks.
Alle 21:21, mercoledì, 12. luglio 2006, hai scritto:
> Vanilla 2.6.16 kernel don't have TLS support, only ones with -bs. Upgrade
> to 2.6.17.
I've got a working system already.
This doesn't change the problem that the kernel + image combination on the
webpage doesn't work out of the box. The kern
13 matches
Mail list logo
