UML (user-mode-linux) kernel-2.6.x
Hi...
Quoted 4 lines
Look at the second item called:
FAQ: "I'm tracing myself and I can't get out"
There is an associated patch which needs to be applied to the UML *guest* when running on a 2.6 host.
It's compliled into the Kernel, but it's still not works at all...
Quoted 2 lines
This patch cant be used with the -mm2 patch
Thx...:
Kurucz "The NeverGone" Istvan :)
==============================================================
--------- Csatlakozz: http://arenaportal.hix.com/ --------- ----- http://arenazo.cjb.net/ -- http://ironiq.hu/aDP/ ----- --- Kurucz "The NeverGone" Istvan: [email protected] --- -------------- http://delfin.klte.hu/~ki0029/ --------------
==============================================================
Quoted 5 lines · The NeverGone
Hi...
Quoted 10 lines
Quoted 6 lines
Look at the second item called:
FAQ: "I'm tracing myself and I can't get out"
There is an associated patch which needs to be applied to the UML *guest* when running on a 2.6 host.It's compliled into the Kernel, but it's still not works at all...
The guest UML kernel??? Works for me. What guest kernel are you running?
Quoted 12 lines
Quoted 7 lines
This patch cant be used with the -mm2 patch
That patch is absolutely not necessary to get basic uml functionality working. It provides the possibility of better performance and more isolation from the host, but it is not needed.
-Tupshin
Quoted 1 line · Tupshin Harper
The guest UML kernel??? Works for me. What guest kernel are you running?
The guest UML using 2.4.23, and it works properly with 2.4.24 host kernel, but not with 2.6.x ...
Quoted 3 lines
That patch is absolutely not necessary to get basic uml functionality working. It provides the possibility of better performance and more isolation from the host, but it is not needed.
So we can't use this patch, but it's not work at all (without this patch, too).
Thx...
The NeverGone :)
==============================================================
--------- Csatlakozz: http://arenaportal.hix.com/ --------- ----- http://arenazo.cjb.net/ -- http://ironiq.hu/aDP/ ----- --- Kurucz "The NeverGone" Istvan: [email protected] --- -------------- http://delfin.klte.hu/~ki0029/ --------------
==============================================================
Quoted 4 lines · The NeverGone
On Sun, 11 Jan 2004, Tupshin Harper wrote:
Quoted 8 lines
Quoted 3 lines
The guest UML kernel??? Works for me. What guest kernel are you running?
The guest UML using 2.4.23, and it works properly with 2.4.24 host kernel, but not with 2.6.x ...
Is that guest 2.4.23 kernel stock, with only the mentioned patch (pasted below) applied?
Are you absolutely positive that you rebuilt that guest kernel with the patch???(I'm guessing that this is really the problem)
Are you still getting the same error message? ("I'm tracing myself...")
-Tupshin
arch/um/os-Linux/process.c Thu Oct 2 14:27:57 2003 renamed from arch/um/os-Linux/process.c Sat Jan 18 12:29:27 2003 +0 -0
@@ -7,6 +7,7 @@
#include <stdio.h>
#include <errno.h>
#include <signal.h>
+#include <linux/unistd.h>
#include <sys/mman.h>
#include <sys/wait.h>
#include "os.h"
@@ -87,7 +88,8 @@
void os_usr1_process(int pid)
{
- kill(pid, SIGUSR1); + syscall(__NR_tkill, pid, SIGUSR1); +/* tkill(pid, SIGUSR1);*/
}
int os_getpid(void)
Quoted 21 lines · Tupshin Harper
Is that guest 2.4.23 kernel stock, with only the mentioned patch (pasted below) applied?
Are you absolutely positive that you rebuilt that guest kernel with the patch???(I'm guessing that this is really the problem)
Are you still getting the same error message? ("I'm tracing myself...")
(quoted diff) +0
-1-Tupshinarch/um/os-Linux/process.c Thu Oct 2 14:27:57 2003 renamed from arch/um/os-Linux/process.c Sat Jan 18 12:29:27 2003 +1
-0@@ -7,6 +7,7 @@ #include <stdio.h> #include <errno.h> #include <signal.h> +#include <linux/unistd.h> #include <sys/mman.h> #include <sys/wait.h> #include "os.h"
arch/um/os-Linux/process.c Thu Oct 2 14:27:57 2003 line 95
Quoted 9 lines
arch/um/os-Linux/process.c Thu Oct 2 14:27:57 2003 renamed from arch/um/os-Linux/process.c Sat Jan 18 12:29:27 2003 +2
-1@@ -87,7 +88,8 @@ void os_usr1_process(int pid) { - kill(pid, SIGUSR1); + syscall(__NR_tkill, pid, SIGUSR1); +/* tkill(pid, SIGUSR1);*/ } int os_getpid(void)
Hi...
It's also included in the host kernel (2.6.1-mm2) ...
Thx...
The NeverGone :)
==============================================================
--------- Csatlakozz: http://arenaportal.hix.com/ --------- ----- http://arenazo.cjb.net/ -- http://ironiq.hu/aDP/ ----- --- Kurucz "The NeverGone" Istvan: [email protected] --- -------------- http://delfin.klte.hu/~ki0029/ --------------
==============================================================
Quoted 18 lines · The NeverGone
Quoted 13 lines · Tupshin Harper
Is that guest 2.4.23 kernel stock, with only the mentioned patch (pasted below) applied?
Are you absolutely positive that you rebuilt that guest kernel with the patch???(I'm guessing that this is really the problem)
Are you still getting the same error message? ("I'm tracing myself...")
-Tupshin
Quoted 10 lines
Hi...
It's also included in the host kernel (2.6.1-mm2) ...
Thx...
The NeverGone :)
You didn't answer any of my questions. I'm forced to believe that the guest is either not properly patched, not properly rebuilt, or not actually the version that you're running. Not much I can suggest other than to recheck those things.
On the odd chance that 2.6.1-mm2 reintroduces a new version of this problem that the mentioned patch doesn't fix, you could try the host running stock 2.6.1.
Also, I inadvertently referred to stock 2.4.23, but of course you would have had to apply a large uml patch to it in order to compile it at all. Did you? If so, which patch, from where?
-Tupshin
Quoted 5 lines · Tupshin Harper
You didn't answer any of my questions. I'm forced to believe that the guest is either not properly patched, not properly rebuilt, or not actually the version that you're running. Not much I can suggest other than to recheck those things.
Quoted 7 lines
On the odd chance that 2.6.1-mm2 reintroduces a new version of this problem that the mentioned patch doesn't fix, you could try the host running stock 2.6.1.
Also, I inadvertently referred to stock 2.4.23, but of course you would have had to apply a large uml patch to it in order to compile it at all. Did you? If so, which patch, from where?
Hi...
But everything works on 2.4.24...!
The NeverGone... :)
==============================================================
--------- Csatlakozz: http://arenaportal.hix.com/ --------- ----- http://arenazo.cjb.net/ -- http://ironiq.hu/aDP/ ----- --- Kurucz "The NeverGone" Istvan: [email protected] --- -------------- http://delfin.klte.hu/~ki0029/ --------------
==============================================================
Quoted 18 lines · The NeverGone
Quoted 5 lines · Tupshin Harper
You didn't answer any of my questions. I'm forced to believe that the guest is either not properly patched, not properly rebuilt, or not actually the version that you're running. Not much I can suggest other than to recheck those things.
Quoted 9 lines
On the odd chance that 2.6.1-mm2 reintroduces a new version of this problem that the mentioned patch doesn't fix, you could try the host running stock 2.6.1.
Also, I inadvertently referred to stock 2.4.23, but of course you would have had to apply a large uml patch to it in order to compile it at all. Did you? If so, which patch, from where?
Quoted 6 lines
Hi...
But everything works on 2.4.24...!
With host 2.4.24, yes. With host 2.6.x, you NEED to apply the patch to the guest. Why is this not clear?
-Tupshin