Here is part of a stack trace I get (by dereferencing a NULL pointer
when a certain assertion failes):
...
Trace: c015b348 <ip_local_deliver+160/1b4>
Trace: c015b56a <ip_rcv+1ce/1fc>
Trace: c0151f5c <net_bh+16c/1cc>
Trace: c0117531 <do_bottom_half+49/64>
Trace: c0107b1d <handle_bottom_half+5/8>
Trace: c010e9d7 <hsfq_setrun+eb/178> ***
Trace: c015d46b <ip_output+6f/a8>
Trace: c015d6f9 <ip_queue_xmit+255/308>
Trace: c0163c83 <tcp_transmit_skb+3e7/3f4>
Trace: c0165404 <tcp_reset_xmit_timer+54/94>
Trace: c010fd32 <wake_up_process+32/80> ***
Trace: c010d80d <send_sig_info+271/280>
Trace: c010dc2e <notify_parent+a2/c4>
Trace: c012e793 <dput+df/148>
...
Note the *** places. Here wake_up_process (not the standard one, from
a different scheduler) calls hsfq_setrun that we see further up the
trace. It does this with interrupts disabled. hsfq_setrun does not
call anything. So the weird thing is that 1. I get extra stack frames
between wake_up_process and hsfq_setrun, and 2. hsfq_setrun appears to
make a function call although (or get interrupted?). Any idea what is
going on?
This is a single processor pII. Kernel 2.2.9 + sfq scheduler. Let me
know if more data is needed.
Thanks,
Borislav
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/