|
Bugzilla – Full Text Bug Listing |
| Summary: | BUG: kernel NULL pointer dereference, address: 0000000d during install on i686 notebook | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE Tumbleweed | Reporter: | Richard Weinberger <richard> |
| Component: | Kernel | Assignee: | openSUSE Kernel Bugs <kernel-bugs> |
| Status: | NEW --- | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | jslaby, mbenes, richard, tiwai |
| Version: | Current | Flags: | jslaby:
needinfo?
(richard) |
| Target Milestone: | --- | ||
| Hardware: | i686 | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Richard Weinberger
2020-07-08 11:38:36 UTC
Is this a regression from the earlier TW releases? Also, any chance to try a newer kernel, e.g. the one in OBS Kernel:stable repo? FWIW, the stack trace implies that file->f_mapping is NULL and crashes at io_is_direct() check. (In reply to Takashi Iwai from comment #1) > Is this a regression from the earlier TW releases? Can't tell. I got recently access to some old notebooks and thought having them run Linux is a good idea. > Also, any chance to try a newer kernel, e.g. the one in OBS Kernel:stable > repo? Since it happens during install it can be a little tricky, but maybe I have time during weekend. Or do you have a pointer to a TW installer with a recent kernel? > FWIW, the stack trace implies that file->f_mapping is NULL and crashes at > io_is_direct() check. FWIW, I see a similar issue on a second Intel Atom (ia32) based Lenovo Notebook, it crashes also while installing. Sadly so hard that I cannot get anything from the console. Thanks, //richard FWIW decoded stacktrace is below. The kernel seems to be a1775d0843b12327f00c719acc05a8224f9ef596. > BUG: kernel NULL pointer dereference, address: 0000000d > #PF: supervisor read access in kernel mode > #PF: error_code(0x0000) - not-present page > *pde = 00000000 > Oops: 0000 [#1] SMP > CPU: 1 PID: 13684 Comm: rpm Not tainted 5.7.5-1-default #1 openSUSE Tumbleweed (unreleased) > Hardware name: Dell Inc. Latitude 2100 /0W785N, BIOS A01 06/02/2009 > EIP: __vfs_write (include/linux/fs.h:3420) > Code: 00 00 8b 73 24 89 55 b0 89 4d b4 89 f0 89 75 ac c1 e8 09 83 e0 02 81 e6 00 40 00 00 0f 85 fd 00 00 00 8b b3 a4 00 00 00 8b 36 <f6> 46 0d 20 0f 85 eb 00 00 00 f7 45 ac 00 10 00 00 75 15 8b b3 a4 > All code > ======== > 0: 00 00 add %al,(%rax) > 2: 8b 73 24 mov 0x24(%rbx),%esi > 5: 89 55 b0 mov %edx,-0x50(%rbp) > 8: 89 4d b4 mov %ecx,-0x4c(%rbp) > b: 89 f0 mov %esi,%eax > d: 89 75 ac mov %esi,-0x54(%rbp) > 10: c1 e8 09 shr $0x9,%eax > 13: 83 e0 02 and $0x2,%eax > 16: 81 e6 00 40 00 00 and $0x4000,%esi > 1c: 0f 85 fd 00 00 00 jne 0x11f > 22: 8b b3 a4 00 00 00 mov 0xa4(%rbx),%esi > 28: 8b 36 mov (%rsi),%esi > 2a:* f6 46 0d 20 testb $0x20,0xd(%rsi) <-- trapping instruction > 2e: 0f 85 eb 00 00 00 jne 0x11f > 34: f7 45 ac 00 10 00 00 testl $0x1000,-0x54(%rbp) > 3b: 75 15 jne 0x52 > 3d: 8b .byte 0x8b > 3e: b3 a4 mov $0xa4,%bl > > Code starting with the faulting instruction > =========================================== > 0: f6 46 0d 20 testb $0x20,0xd(%rsi) > 4: 0f 85 eb 00 00 00 jne 0xf5 > a: f7 45 ac 00 10 00 00 testl $0x1000,-0x54(%rbp) > 11: 75 15 jne 0x28 > 13: 8b .byte 0x8b > 14: b3 a4 mov $0xa4,%bl > EAX: 00000000 EBX: ddb3a180 ECX: 00005e44 EDX: bfd3af6c > ESI: 00000000 EDI: c28fdf7c EBP: c28fdf4c ESP: c28fdef4 > DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 EFLAGS: 00010246 > CR0: 80050033 CR2: 0000000d CR3: 32744000 CR4: 000006d0 > Call Trace: > vfs_write (fs/read_write.c:560) > ksys_write (fs/read_write.c:613) > __ia32_sys_write (fs/read_write.c:621) > do_int80_syscall_32 (arch/x86/entry/common.c:337) > entry_INT80_32 (arch/x86/entry/entry_32.S:1085) Richard, have you had a chance to try a new installation image? Some time has passed since the report, so it would be useful to retry. It may have been fixed meanwhile. TW is on 5.9.x kernel now. (In reply to Miroslav Beneš from comment #4) > Richard, have you had a chance to try a new installation image? Some time > has passed since the report, so it would be useful to retry. It may have > been fixed meanwhile. TW is on 5.9.x kernel now. Thanks for getting back to me! I gave it a try but the installation failed, sadly without output. While unpacking packages the device stopped and rebooted. :-( It appears to me as HW failures. Are the crashes always the same or they vary? Anyway, this is 32bit, so it won't draw much attention :/. |