From Bugzilla Helper: User-Agent: Mozilla/5.0 (compatible; Konqueror/3.1; Linux) Description of problem: When I attach the firewire device, the hotplug script loads the modules as it should, but after loading sbp2, it dumps a message (addresses vary depending on kernel ver/patch) Platform dell X200 + firewire CDRW/DVD Version-Release number of selected component (if applicable): kernels 2.4.21 and 2.6.0 test1/2/2+mm/2+bk5 How reproducible: Always Steps to Reproduce: 1. compile the ieee1394 module 2. install the new configuration 3. reboot Actual Results: Slab corruption: start=d552d8c0, expend=d552d937, problemat=d552d900 Last user: [<f891d1fd>](free_hpsb_packet+0x2c/0x33 [ieee1394]) Data: ****************************************************************6A ******************************************************A5 Next: 71 F0 2C .FD D1 91 F8 71 F0 2C ..................... slab error in check_poison_obj(): cache `hpsb_packet': object was modified after freeing Call Trace: [<c014e5b1>] check_poison_obj+0x16c/0x1ac [<c014e77d>] slab_destroy+0x18c/0x194 [<c0152532>] cache_reap+0x2d6/0x45c [<c01515ef>] reap_timer_fnc+0x0/0x29 [<c01515fa>] reap_timer_fnc+0xb/0x29 [<c012df3b>] update_wall_time+0xf/0x3a [<c012e26c>] run_timer_softirq+0x18b/0x429 [<c01127e7>] timer_interrupt+0x17f/0x3be [<c0129051>] do_softirq+0xa1/0xa3 [<c010cc2f>] do_IRQ+0x256/0x3a4 [<c0124dc7>] profile_hook+0x2f/0x4d [<c0107269>] default_idle+0x0/0x2c [<c0107269>] default_idle+0x0/0x2c [<c010ab70>] common_interrupt+0x18/0x20 [<c0107269>] default_idle+0x0/0x2c [<c0107269>] default_idle+0x0/0x2c [<c0107290>] default_idle+0x27/0x2c [<c0107301>] cpu_idle+0x31/0x3a [<c0105000>] rest_init+0x0/0xf5 [<c0398725>] start_kernel+0x1b8/0x210 [<c039843f>] unknown_bootoption+0x0/0xfa Expected Results: sbp2 should load normally and subsequent modules/scripts should be able to use the firewire device. Additional info: The system is usable, but sbp2 cannot be removed and subsequently, no attempts can be made to fix it.
Matt, have you tested Severn with sbp2 on your setup?
Is this repeatable with the sbp2 module from the kernel on the Red Hat CD ? Your report doesn't specifically mention any Red Hat kernels. If you are only seeing it with upstream kernels, please report the bug to the upstream sbp2 maintainer.
The bug is reproductible with the 2.4.21 nptl kernel delivered on the severn's ISO.
Created attachment 93688 [details] System log from 2.6.0 Notice numerous references that firewire is detected.
Ditto here. I have a Buslink 48x12x48x and it works great in ol Shrike. Severn though with 2.6.0. See attached log above. I will also attach a copy of lspci.
Created attachment 93689 [details] lspci More info for your viewing pleasure
Created attachment 93690 [details] Kernel Startup Log More viewing enjoyment to help with Severn and 2.6.0
2.6.0 test4 kernel made no difference in outcome. I made all the updates on 9.0.93 to no avail. IEEE1394 is dead here. I might add that usb-ohci is also hosed. Raxet
I posted this to the linux1394-devel mailing list and the suggestion for this stack trace was: 1) fix the poisoned slab 2) turn of kernel debugging as a work around
FWIW, I can't duplicate this with 2.4.22-20.1.2024.2.36.nptl
*** Bug 100545 has been marked as a duplicate of this bug. ***
*** This bug has been marked as a duplicate of 103821 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.