Red Hat Bugzilla – Bug 127279
kernel panic in try_to_identify() -> prob_irq_off()
Last modified: 2013-03-06 00:57:07 EST
From Bugzilla Helper: User-Agent: Mozilla/5.0 (compatible; Konqueror/3.2; Linux) (KHTML, like Gecko) Description of problem: In testing for bug #127105, we set inittab to boot runlevel 6 (reboot) in order to stress test the boot code on an x445 and trigger the problem. However, I keep on running into what I believe to be a different problem w/ the IDE layer. This issue still appears after the workaround for bug #127105 is applied to a custom kernel and does not appear when the IDE subsystem is configured out of the kernel. Its not every time I see this, but usually within an hour or so I'll get the following panic: EIP: 0010:[<c0109371>] Not tainted EFLAGS: 00010217 EIP is at probe_irq_off [kernel] 0x21 eax: 00000005 ebx: c0358800 ecx: 00000190 edx: 1d078075 esi: 00000000 edi: 00000000 ebp: 00000000 esp: f7f65ecc ds: 0018 es: 0018 ss: 0018 Process swapper (pid: 1, stackpage=f7f65000) Stack: c0358804 c041bc60 fffffef8 00000001 00000002 c01b9a50 fffffef8 ecffffef c041bc20 c041bc60 000000ec c041bc20 c01b9b73 c041bc60 000000ec ec1b9e86 c041bc60 c041bd58 c041bc20 00000000 c01ba0e8 c041bc60 000000ec c041bd58 Call Trace: [<c01b9a50>] try_to_identify [kernel] 0x90 (0xf7f65ee0) [<c01b9b73>] do_probe [kernel] 0xe3 (0xf7f65efc) [<c01ba0e8>] probe_hwif [kernel] 0xe8 (0xf7f65f1c) [<c01bac6d>] ideprobe_init [kernel] 0x5d (0xf7f65f50) [<c02569c0>] call_spurious_interrupt [kernel] 0xf23f (0xf7f65fb0) [<c0105000>] stext [kernel] 0x0 (0xf7f65fcc) [<c0105278>] init [kernel] 0x28 (0xf7f65fdc) [<c0105000>] stext [kernel] 0x0 (0xf7f65fe8) [<c0105856>] arch_kernel_thread [kernel] 0x26 (0xf7f65ff0) [<c0105250>] init [kernel] 0x0 (0xf7f65ff8) Code: 00 30 43 fc 7f 88 44 fe 12 00 8b 03 a9 10 00 00 00 74 1c a9 Version-Release number of selected component (if applicable): kernel-2.4.9-e.40-summit How reproducible: Sometimes Steps to Reproduce: 1. Set inittab to default to runlevel 6 2. Reboot the system 3. Watch as the system reboots itself over and over. Actual Results: Within an our or so, the panic above usually appears Expected Results: No problems, the system continues rebooting over and over. Additional info:
No Bob, that would be U6.....
Wendy, please open an IT for this one, thanks.
Bob, Chris has opened Issue #47021.
Reproduced a very similar panic doing the same reboot stress test w/ RHEL3 U4 beta1 EIP: 0010:[<c041dc20>] Not tainted EFLAGS: 00010246 EIP is at ide_hwifs [kernel] 0x0 eax: 00000000 ebx: c035b880 ecx: 00000000 edx: 00000001 esi: c041dc60 edi: 00000000 ebp: 00000000 esp: c3275ee4 ds: 0018 es: 0018 ss: 0018 Process swapper (pid: 1, stackpage=c3275000) Stack: c01bab14 ecffffef c041dc20 c041dc60 000000ec c041dc20 c01bac93 c041dc60 000000ec ec1bafa6 c041dc60 c041dd58 c041dc20 00000000 c01bb208 c041dc60 000000ec c041dd58 00000246 000000dc c041df64 00000000 00000000 00000000 Call Trace: [<c01bab14>] try_to_identify [kernel] 0x34 (0xc3275ee4) [<c01bac93>] do_probe [kernel] 0xe3 (0xc3275efc) [<c01bb208>] probe_hwif [kernel] 0xe8 (0xc3275f1c) [<c01bbd8d>] ideprobe_init [kernel] 0x5d (0xc3275f50) [<c0257bc0>] call_spurious_interrupt [kernel] 0xf33f (0xc3275fb0) [<c0105000>] stext [kernel] 0x0 (0xc3275fcc) [<c0105278>] init [kernel] 0x28 (0xc3275fdc) [<c0105000>] stext [kernel] 0x0 (0xc3275fe8) [<c0105856>] arch_kernel_thread [kernel] 0x26 (0xc3275ff0) [<c0105250>] init [kernel] 0x0 (0xc3275ff8) Code: 00 00 00 00 00 00 00 00 f0 01 f1 01 f2 01 f3 01 f4 01 f5 01 <0>Kernel panic: not continuing
Ack, sorry! I didn't mean RHEL3 U4 beta1, I meant RHAS 2.1 U6 beta1.
John, was this reproduced using BIOS 45A (ver 1.17)?
I'm working to reproduce the issue w/ the updated BIOS (v1.17).
It appears the new BIOS (v1.17) has resolved this issue.