![]() ![]() Ohci_hcd 0000:c8:01.0: OHCI Host Controller Ohci_hcd 0000:c8:01.1: USB bus 3 deregistered Ohci_hcd 0000:c8:01.0: USB bus 2 deregistered It takes approximately half a hour to fail into xmon, I=0 while : do echo $((i++)) rmmod ohci-hcd modprobe ohci-hcd done ![]() It looks like the problem can be reproduced by load and unload ohci-hcd module Mi show information about memory allocation La lookup symbol+offset of specified address The following are some outputs within xmon,Ĭ# try to switch to cpu number h (in hex) ![]() With Shell command trace on, It is actually failing in /etc/rc.sysinit here,Ĭpu 0x5: Vector: 300 (Data Access) at Like the bug is introduced by new patches. Reboots each, and I can still reproduce this reliably on new kernel, so it looks I could not reproduce this bug on both -67.EL and -67.0.7.EL kernel with 700 There are a lot of irq validity checks in upstream code, e.g.: Get_irq_desc() (which free_irq() calls), have no any checks: As I've understood thatĭata access fault, the irq argument for free_irq() was out of bounds and I think this is a new bug in IBM's EHEA network adapter. Version-Release number of selected component (if applicable):īoth 2.6.9-67.0.12.EL and 2.6.9-67.0.13.EL kernel I have seen the failure 3 times in around 700 reboots occurred only on two I could not capture a vmcore, as it failed before start of netdump daemon. storage network audiocpu 0x0: Vector: 300 (Data While booting the latest RHEL4U6 Z-stream kernel on ppc64 machines, it failed ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |