No, not a HyperThreaded Pentium 4 Dual Xeon Extreme Edition(TM)... On a
lark, I decided to build a Pentium 166 (with a whopping 32MB RAM) system
from spare parts and boot a development version of Finnix
(www.finnix.org) on it. After Finnix was up, I decided to give UML a
try. This was the re
i think everytime you load it and unload it you may get different addresses
so you want a few more things in the [Host](gdb) section before loading
a module these stepd should allow you to debug a module_init routine
1) from http://kgdb.linsyssoft.com/downloads/miscmacros
we get an idea how t
On Fri, Mar 24, 2006 at 10:21:27AM +0700, Mikado wrote:
> I 'chattr +i' /mnt/a and mount (using hostfs) it on UML /mnt/b then
> 'lsattr' /mnt/b on UML will panic the kernel.
I always love a clear way to reproduce a bug.
However, with the new externfs-based hostfs, this doesn't crash. I
get no ou
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
- - How can I debug loadable modules at run time?
Here is my way:
( /tmp/mymodule.ko exists on both [Host] and [UML] )
+ [Host] # gdb linux
+ [Host] (gdb) ...
+ [Host] (gdb) r
+ [UML] # insmod /tmp/mymodule.ko
+ [UML] # cat /sys/module/
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I 'chattr +i' /mnt/a and mount (using hostfs) it on UML /mnt/b then
'lsattr' /mnt/b on UML will panic the kernel.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFE
> In 2.6.16, the kernel allows user-selectable address space splits. I
> have 1.5G of memory, so choosing 2G/2G lets me turn off highmem. I also
> have the new 2.6.16 skas3 v8.2 patch applied.
>
> With this configuration, all umls die immediately, whether I try skas0,
> skas3 or TT.
>
> Is this
On Fri, Mar 24, 2006 at 01:56:32AM +0100, Blaisorblade wrote:
> I don't think it's easy - however, it can be done. Below my idea - Jeff,
> please comment on this
>
> In assembly.
> It goes like this:
>
> call label /*this is coded like a relative jump, i.e. call +1*/
> label: pop $eax /*we got t
On Friday 24 March 2006 01:40, Jason Lunz wrote:
> On Thu, Mar 23, 2006 at 06:50:44PM -0500, Jeff Dike wrote:
> > You can make some progress by disabling CONFIG_MODE_TT - that will
> > cause UML to load at a normal address rather than the 0xa000 that
> > CONFIG_MODE_TT will force.
> >
> > That
On Thursday 23 March 2006 15:50, Serge Goodenko wrote:
> > In my experience, it stops on some breakpoints and not on others (for
> > instance it doesn't break on panic). You may see if attaching to an
> > already running UML rather than starting it under GDB makes a difference
> > ("gdb kernelBinar
On Thu, Mar 23, 2006 at 06:50:44PM -0500, Jeff Dike wrote:
> You can make some progress by disabling CONFIG_MODE_TT - that will
> cause UML to load at a normal address rather than the 0xa000 that
> CONFIG_MODE_TT will force.
>
> That should get you to the point of trying to run init.
>
> Sinc
On Thu, Mar 23, 2006 at 10:13:47PM +, Jason Lunz wrote:
> In 2.6.16, the kernel allows user-selectable address space splits. I
> have 1.5G of memory, so choosing 2G/2G lets me turn off highmem. I also
> have the new 2.6.16 skas3 v8.2 patch applied.
>
> With this configuration, all umls die im
On Thursday 23 March 2006 08:15, David Lang wrote:
> On Thu, 23 Mar 2006, Blaisorblade wrote:
> > On Thursday 23 March 2006 00:55, David Lang wrote:
> > That's _required_ but not enough; to run a 32-bit UML binary you _need_
> > that support on the host.
> yep, I didn't realize it was missing bef
On Thursday 23 March 2006 10:57, Florian Effenberger wrote:
> Hi,
> >> eth0 IP: 85.123.123.10
> > Let's say "IP A"
> >> eth0:1 IP: 85.123.123.15
> > "IP B"
> > That could be due to a network hang fixed _after_ 2.4.25 - upgrade to my
> > 2.4.x -bs2 patchset on my homepage.
> I run 2.6.8. :-)
On Fri, 24 Mar 2006, Blaisorblade wrote:
On Thursday 23 March 2006 04:34, David Lang wrote:
On Thu, 23 Mar 2006, Blaisorblade wrote:
On Thursday 23 March 2006 02:09, Nix wrote:
On Wed, 22 Mar 2006, David Lang uttered the following:
Debian there is trying to be smart anyway and helping the us
On Thursday 23 March 2006 04:34, David Lang wrote:
> On Thu, 23 Mar 2006, Blaisorblade wrote:
> > On Thursday 23 March 2006 02:09, Nix wrote:
> >> On Wed, 22 Mar 2006, David Lang uttered the following:
> > Debian there is trying to be smart anyway and helping the user which
> > forgot "make" (and p
On Thu, Mar 23, 2006 at 02:19:16PM -0800, David Lang wrote:
> I don't know about other non-standard memory configs, but there is an
> explicit config option when compiling a uml kernel to support the 2G/2G
> memory split.
I see. Do you know whether a single uml binary can support arbitrary
host
On Thu, 23 Mar 2006, Jason Lunz wrote:
In 2.6.16, the kernel allows user-selectable address space splits. I
have 1.5G of memory, so choosing 2G/2G lets me turn off highmem. I also
have the new 2.6.16 skas3 v8.2 patch applied.
With this configuration, all umls die immediately, whether I try ska
In 2.6.16, the kernel allows user-selectable address space splits. I
have 1.5G of memory, so choosing 2G/2G lets me turn off highmem. I also
have the new 2.6.16 skas3 v8.2 patch applied.
With this configuration, all umls die immediately, whether I try skas0,
skas3 or TT.
Is this something uml ca
On Thu, 23 Mar 2006, David Lang wrote:
On Wed, 22 Mar 2006, David Lang wrote:
Btw, you can also download pre-built UML binaries from my homepage (see
signature), and this is useful even to determine what's going on (e.g. say
that binary work, your one doesn't, let's compare configurations! Or
On Wed, 22 Mar 2006, David Lang wrote:
Btw, you can also download pre-built UML binaries from my homepage (see
signature), and this is useful even to determine what's going on (e.g. say
that binary work, your one doesn't, let's compare configurations! Or a
couple
of months ago the answer was "
>
> on /etc/inittab:
>
> c0:1235:respawn:/sbin/getty 38400 tty0 linux
>
> on command line:
>
> user$ linux ubd0=... con0=fd:0 con=pty ...
>
> In this way you have a console, if you don't specify con=pty, the default
> option is the xterm.
> This is to say that it doesn't matter the devfs, if
On Thu, Mar 23, 2006 at 10:40:17AM -0500, Lull, Rick wrote:
Hi Rick,>
> My previous UML hosts were 2.4 or 2.6.8 kernels, so they had DevFS.
> I noted the announcements that udev was the replacement for same and it was
> going to happen sometime in the 2.6 train of releases. Now, with 2.6.15
Have you reviewed the wiki? They have at least one page concerning this
at:
http://uml.harlowhill.com/index.php/DevfsUdev
Tony
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Lull, Rick
> Sent: Thursday, March 23, 2006 7:40 AM
> To: 'user-mode
Hi folks -
Whilst setting up a test machine to be able to test some scripts and
not blow anything up on the production side, I've run into a few snags.
My previous UML hosts were 2.4 or 2.6.8 kernels, so they had DevFS.
I noted the announcements that udev was the replacement for s
> In my experience, it stops on some breakpoints and not on others (for
> instance
> it doesn't break on panic). You may see if attaching to an already running
> UML rather than starting it under GDB makes a difference ("gdb kernelBinary"
> and "attach ", where PID is the lowest one between th
Hi,
thanks for your mail.
eth0 IP: 85.123.123.10
Let's say "IP A"
eth0:1 IP: 85.123.123.15
"IP B"
That could be due to a network hang fixed _after_ 2.4.25 - upgrade to my 2.4.x
-bs2 patchset on my homepage.
I run 2.6.8. :-)
host: tap0 uses "IP A", eth0:1 is removed.
UML: eth0 is as
26 matches
Mail list logo
