Kernel BUG, 2.6.16 and higher[Adaptec 29320]
Sean Bruno
sean.bruno at dsl-only.net
Fri Aug 11 04:40:15 UTC 2006
I posted a BUG notification on lkml and just thought I would throw it
over here as well.
Between 2.6.15 and 2.6.16 there was quite a bit of work to the aic79xx
driver which seems to have inflicted a nasty bug when an external array
is powered on.
Scenario:
1. System running 2.6.16 or higher is up and running with an external
SCSI array attached in a powered off state.
2. Power up the array.
3. Linux system locks up hard with the following BUG displayed on the
console:
scsi0: Someone reset channel A
BUG: soft lockup detected on CPU#0!
<c013258b> softlockup_tick+0x7f/0x8e <c011e6e4> update_process_times
+0x35/0x57 <c0104eda> timer_interrupt+0x3d/0x60 <c0132698>
handle_IRQ_event+0x21/0x4a
<c0132739> __do_IRQ+0x78/0xcb <c0103eb5> do_IRQ+0x6b/0x7a
<c0102c4a> common_interrupt+0x1a/0x20 <c02d3a2a>
_spin_unlock_irqrestore+0xa/c <e08d0481> ahd_linux_isr+0x173/0x17f
[aic79xx] <c0132698> handle_IRQ_event+0xa <c0132739> __do_IRQ+0x78/0xcb
<c0103ea8> do_IRQ+0x5e/0x7a
=======================
<c0102c4a> common_interrupt+0x1a/0x20 <c011b5bc> __do_softirq
+0x2c/0x7d
<c0103f8c> do_softirq+0x38/0x3f
=======================
<c0103eba> do_IRQ+0x70/0x7a <c0102c4a> common_interrupt+0x1a/0x20
<c0101150> mwait_idle+0x1a/0x2a <c01010bf> cpu_idle+0x40/0x5c
<c038c618> start_kernel+0x184/0x186
I have forwarded this information to the folks who posted the last
changes to the driver in 2.6.16
If anyone else has the time/inclination/hardware, can they verify this
for me?
Sean
More information about the fedora-list
mailing list