2.1.93 died on my SMP PPro 64MB /ide drive, with the following
message being sent to my serial console (copied down by hand)
(the message below was printed twice)
wait on irq, cpu 1
irq 0 [0 0]
bh 1 [1 0]
then the following address's
c016efd? c016cfe5 c017b608
c015a44f c0134614 c0134a29
which from System.map translate to =>
tcp_ack_saw_tstamp <c016efdc>
tcp_poll <c016cf48>
inet_poll <c017b5e4>
sock_poll <c015a430>
do_select <c013448c>
sys_select <c0134704>
using magic sysrq to display the pc showed a couple of different
pc's when I pressed it a couple of times the places of note where
tcp_sltimer_handler
do_timer
now using magic sysrq again to show memory revealed that
there were
Total network buffer allocations : 6258759
but this machine is just my desktop machine, the network card isn't
connected to anything (+2.1.93 broke the network drivers), + there was no
ppp link or anything. The uptime had been ~ 2 days. Crashes like this have
been occuring in X, since around 2.1.90, if the machine is left idle for a
long time (~10 hours or more)
-- eman
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu