Mailinglist Archive: opensuse-bugs (6131 mails)
| < Previous | Next > |
[Bug 293977] New: kernel-debug fails in early boot
- From: bugzilla_noreply@xxxxxxxxxx
- Date: Mon, 23 Jul 2007 14:08:24 -0600 (MDT)
- Message-id: <bug-293977-21960@xxxxxxxxxxxxxxxxxxxxxxxxx/>
https://bugzilla.novell.com/show_bug.cgi?id=293977
Summary: kernel-debug fails in early boot
Product: openSUSE 10.2
Version: Final
Platform: i686
OS/Version: openSUSE 10.2
Status: NEW
Severity: Normal
Priority: P5 - None
Component: Kernel
AssignedTo: kernel-maintainers@xxxxxxxxxxxxxxxxxxxxxx
ReportedBy: Chris.Horler@xxxxxxxxxxx
QAContact: qa@xxxxxxx
Found By: Other
Created an attachment (id=152499)
--> (https://bugzilla.novell.com/attachment.cgi?id=152499)
GDB disassembly of __alloc_bootmem_core
I'm helping out with some driver development for the linux UVC driver, it would
be nice for me to have a working debug kernel. Currently it doesn't even get
far enough to write a proper Oops for me
(Note the standard Suse kernel works fine - so I suspect a bug in the
additional debug routines in kernel-debug, but can't understand why no one else
has this problem).
I have duplicated the following from my camera - I have no serial console.
Unknown interrupt or fault at EIP 00000060 c04420e5 00000002
Unknown interrupt or fault at EIP 00000060 c01002b1 000020e4
Unknown interrupt or fault at EIP 00000060 c01002b1 000002b0
Unknown interrupt or fault at EIP 00000060 c01002b1 000002b0
.. <many more identical to the last two lines and then the machine reboots>
c01002b1 does not appear to be a valid address according to gdb.
c04420e5 seems to be a call to memset (__constant_c_memset) in
__alloc_bootmem_core.
$ gdb /boot/vmlinux-2.6.18.8-0.3-debug
..
(gdb) l *0xc04420e5
0xc04420e5 is in __alloc_bootmem_core (include/asm/string.h:372).
367 * area at compile-time..
368 */
369 static __always_inline void * __constant_c_memset(void * s, unsigned
long c, size_t count)
370 {
371 int d0, d1;
372 __asm__ __volatile__(
373 "rep ; stosl\n\t"
374 "testb $2,%b3\n\t"
375 "je 1f\n\t"
376 "stosw\n"
I don't know exactly what else to do to help solve this - but I'm capable of
helping if required (using and developing on Linux now for ~9 years).
A disassembly of __alloc_bootmem_core is attached.
Regards,
Chris
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
Summary: kernel-debug fails in early boot
Product: openSUSE 10.2
Version: Final
Platform: i686
OS/Version: openSUSE 10.2
Status: NEW
Severity: Normal
Priority: P5 - None
Component: Kernel
AssignedTo: kernel-maintainers@xxxxxxxxxxxxxxxxxxxxxx
ReportedBy: Chris.Horler@xxxxxxxxxxx
QAContact: qa@xxxxxxx
Found By: Other
Created an attachment (id=152499)
--> (https://bugzilla.novell.com/attachment.cgi?id=152499)
GDB disassembly of __alloc_bootmem_core
I'm helping out with some driver development for the linux UVC driver, it would
be nice for me to have a working debug kernel. Currently it doesn't even get
far enough to write a proper Oops for me
(Note the standard Suse kernel works fine - so I suspect a bug in the
additional debug routines in kernel-debug, but can't understand why no one else
has this problem).
I have duplicated the following from my camera - I have no serial console.
Unknown interrupt or fault at EIP 00000060 c04420e5 00000002
Unknown interrupt or fault at EIP 00000060 c01002b1 000020e4
Unknown interrupt or fault at EIP 00000060 c01002b1 000002b0
Unknown interrupt or fault at EIP 00000060 c01002b1 000002b0
.. <many more identical to the last two lines and then the machine reboots>
c01002b1 does not appear to be a valid address according to gdb.
c04420e5 seems to be a call to memset (__constant_c_memset) in
__alloc_bootmem_core.
$ gdb /boot/vmlinux-2.6.18.8-0.3-debug
..
(gdb) l *0xc04420e5
0xc04420e5 is in __alloc_bootmem_core (include/asm/string.h:372).
367 * area at compile-time..
368 */
369 static __always_inline void * __constant_c_memset(void * s, unsigned
long c, size_t count)
370 {
371 int d0, d1;
372 __asm__ __volatile__(
373 "rep ; stosl\n\t"
374 "testb $2,%b3\n\t"
375 "je 1f\n\t"
376 "stosw\n"
I don't know exactly what else to do to help solve this - but I'm capable of
helping if required (using and developing on Linux now for ~9 years).
A disassembly of __alloc_bootmem_core is attached.
Regards,
Chris
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
| < Previous | Next > |