rmmod ohci1394 hangs

Ben,

this sequence of commands will hang the final rmmod.

init=/bin/bash --login
mount proc
modprove -v ohci1394
rmmod ohci1394
modprobe -v ohci1394
rmmod ohci1394 -> hangs

It is stuck here:

rmmod D 0FF7D030 0 4518 1 4517 (NOTLB)

Call trace:
 [c000a188] __switch_to+0x48/0x70
 [c0171588] schedule+0x2b0/0x5d0
 [c0171b28] wait_for_completion+0x7c/0xec
 [e90492fc] nodemgr_remove_host+0x70/0x108 [ieee1394]
 [e904606c] __unregister_host+0xd8/0x10c [ieee1394]
 [e9046170] highlevel_remove_host+0x9c/0x15c [ieee1394]
 [e9044e7c] hpsb_remove_host+0x8c/0xdc [ieee1394]
 [e9065c80] ohci1394_pci_remove+0xd8/0x418 [ohci1394]
 [c00b487c] pci_device_remove+0x60/0x64
 [c00e8a68] device_release_driver+0x84/0x88
 [c00e8b1c] bus_remove_driver+0xb0/0x12c
 [c00e9270] driver_unregister+0x1c/0xa0
 [c00b4ac4] pci_unregister_driver+0x20/0x88
 [e9068698] ohci1394_cleanup+0x44/0x1054 [ohci1394]
 [c00352f8] sys_delete_module+0x1d0/0x2a8

knodemgrd_0 exits on the first rmmod, but leaves nodemgr_serialize in down state. This patch fixes it for me.

diff -purN linux-2.6.9-rc2-bk6.orig/drivers/ieee1394/nodemgr.c linux-2.6.9-rc2-bk6/drivers/ieee1394/nodemgr.c
@@ -1490,7 +1490,10 @@ static int nodemgr_host_thread(void *__h
 		}
 
 		if (hi->kill_me)
+		{
+			up(&nodemgr_serialize);
 			break;
+		}
 
 		/* Pause for 1/4 second in 1/16 second intervals,
 		 * to make sure things settle down. */

@@ -1515,7 +1518,10 @@ static int nodemgr_host_thread(void *__h
 
 			/* Check the kill_me again */
 			if (hi->kill_me)
+			{
+				up(&nodemgr_serialize);
 				goto caught_signal;
+			}
 		}
 
 		if (!nodemgr_check_irm_capability(host, reset_cycles)) {

USB is for mice, FireWire is for men!

sUse lINUX ag, nÜRNBERG

Ben,

sw-suspend does not work with our kernel.
khpsbpkt will die because down_interruptible returns -EINTR on suspend. As a result, hpsb packet delivery will not work anymore after resume.

knodemgrd_0 D 00000000 0 5291 1 5170 (L-TLB)
c30cbe98 00000046 00000010 00000000 ab7b5340 000f4221 c0c54ae0 c5490aa0

       c5490bf0 c30ca000 c30cbf04 c30cbeb4 c30cbeec c03053be 00000000 c5490aa0 
       c0119c00 00000000 00000000 c0c54aec c6e08000 c89e758b 00000001 c5490aa0 
Call Trace:
 [<c03053be>] wait_for_completion+0x6e/0xa0
 [<c0119c00>] default_wake_function+0x0/0x10
 [<c89e758b>] hpsb_packet_received+0x6b/0x70 [ieee1394]
 [<c0119c00>] default_wake_function+0x0/0x10
 [<c89e6c99>] hpsb_send_packet_and_wait+0x49/0x60 [ieee1394]
 [<c89e7dad>] hpsb_make_readpacket+0x6d/0x80 [ieee1394]
 [<c89e8189>] hpsb_read+0x49/0xd0 [ieee1394]
 [<c89e81a1>] hpsb_read+0x61/0xd0 [ieee1394]
 [<c89eb953>] nodemgr_bus_read+0x53/0x90 [ieee1394]
 [<c89f05f0>] csr1212_parse_bus_info_block+0xb0/0x130 [ieee1394]
 [<c89f0d9e>] csr1212_parse_csr+0x1e/0x90 [ieee1394]
 [<c89ed353>] nodemgr_node_scan_one+0x63/0x1e0 [ieee1394]
 [<c89ed529>] nodemgr_node_scan+0x59/0x60 [ieee1394]
 [<c89edba4>] nodemgr_host_thread+0x144/0x180 [ieee1394]
 [<c89eda60>] nodemgr_host_thread+0x0/0x180 [ieee1394]
 [<c0104255>] kernel_thread_helper+0x5/0x10

diff -purN linux-2.6.9-rc3/drivers/ieee1394/ieee1394_core.c linux-2.6.9-rc3.bug45851/drivers/ieee1394/ieee1394_core.c
@@ -1030,15 +1030,19 @@ static int hpsbpkt_thread(void *__hi)
 
 	daemonize("khpsbpkt");
 
-	while (!down_interruptible(&khpsbpkt_sig)) {
-		if (khpsbpkt_kill)
+	while (1) {
+		if (down_interruptible(&khpsbpkt_sig)) {
+			if (current->flags & PF_FREEZE) {
+				refrigerator(0);
+				continue;
+			}
+			printk("khpsbpkt: received unexpected signal?!\n" );
 			break;
-
-		if (current->flags & PF_FREEZE) {
-			refrigerator(0);
-			continue;
 		}
 
+		if (khpsbpkt_kill)
+			break;
+
 		while ((skb = skb_dequeue(&hpsbpkt_queue)) != NULL) {
 			packet = (struct hpsb_packet *)skb->data;
 

USB is for mice, FireWire is for men!

sUse lINUX ag, nÜRNBERG

Hi.

I've applied the fix to the suspend2 tree fwiw.

Nigel

Nigel Cunningham
Pastoral Worker
Christian Reformed Church of Tuggeranong
PO Box 1004, Tuggeranong, ACT 2901

Many today claim to be tolerant. True tolerance, however, can cope with others
being intolerant.