Created attachment 333774 [details] The dmesg pre-captured before crash Description of problem: After pairing the earpiece, kernel oopses (GPF). Version-Release number of selected component (if applicable): 2.6.29-0.148.rc6.fc11.x86_64 How reproducible: 100% (in the given configuration) Steps to Reproduce: 1. Connect USB adapter 2. hcitool scan 3. hcitool cc MAC 4. Wait - it's not synchronous. Takes some idle time to happen. 5. Oops Actual results: Crash Expected results: No crash at least Additional info: Terminal trace: [root@simbelmyne zaitcev]# hcitool dev Devices: hci0 00:18:E7:36:30:3A [root@simbelmyne zaitcev]# hcitool dev Devices: hci0 00:18:E7:36:30:3A [root@simbelmyne zaitcev]# hcitool scan Scanning ... 00:1A:0E:43:6D:5E Motorola H350 [root@simbelmyne zaitcev]# hcitool cc 00:1A:0E:43:6D:5E [root@simbelmyne zaitcev]# hcitool cc 00:1A:0E:43:6D:5E [root@simbelmyne zaitcev]# <--------------- crash at this point after some idle time I've done hcitool cc twice here, but actually it happens with just one too.
Created attachment 333775 [details] The crash, over netconsole, same run as dmesg
0: 49 8b 0c 24 mov (%r12),%rcx r12 == 6b6b6b6b6b6b6b6b
*** Bug 481678 has been marked as a duplicate of this bug. ***
net/core/skbuff.c:1793: result = __skb_dequeue(list); static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list) { struct sk_buff *skb = skb_peek(list); if (skb) __skb_unlink(skb, list); return skb; } skb_peek() returns skb == 0x6b6b6b6b6b6b6b6b static inline struct sk_buff *skb_peek(struct sk_buff_head *list_) { struct sk_buff *list = ((struct sk_buff *)list_)->next; if (list == (struct sk_buff *)list_) list = NULL; return list; }
reported upstream: http://marc.info/?l=linux-netdev&m=123612024109003&w=2
Could this be related as well? http://thread.gmane.org/gmane.linux.bluez.kernel/1549
The bluetooth patches have been rebased in kernel 0.207, can you try this again?
2.6.29-0.207.rc7.fc11 seems to work ok (I've not gotten the headset to work, but there's no crash anymore). Rawhide is on .197, I pulled .207 from Koji to test. Closing.
I've seen some lockups after finishing with Bluetooth on kernel-2.6.29-3.fc10.x86_64, have the patches made it into this kernel?