Always happens when i plug a usb stick through the usb hub. Often it happens when i boot Fedora first, so i boot Gentoo and i get the panic. Happened once when i plugged a usb stick in the motherboard front port. Don't know if related but once happened also at boot time after the message "/dev already mounted" and before starting udevd but with a different message: "BUG: unable to handle kernel NULL pointer dereference at 00000000....040 et cetera. As usual: comm:ksoftirqd/1 Not tainted.... Full messages in the attached file (shoot with a camera). Reproducible: Always Steps to Reproduce: 1.Plug usb stick in usb hub 1a.Reboot from Fedora 1b.Plug usb stick Actual Results: Kernel panic[...]Pid: X, comm: ksoftirqd/Y Tainted/Not tainted G D C 2.6.38-gentoo-r6-unosd #Z (System manufacturer System Product Name/M4A88TD-V EVO/USB3 Expected Results: Don't panic! Hw: Athlon II X4 640 4GB RAM DDR2 Motherboard ASUS M4A88TD-V EVO/USB3 Attached jpg log, kernel's .config.
Created attachment 279755 [details] First time, at boot, after rebooting from Fedora
Created attachment 279757 [details] Second time, at boot, before udevd.
Created attachment 279759 [details] Third time, after plugging usb stick in front motherboard port.
Created attachment 279761 [details] My kernel configuration
Have you tried with latest 2.6.39 kernel available in testing?
(In reply to comment #5) > Have you tried with latest 2.6.39 kernel available in testing? Not yet. But i do prefer stable kernel. (Even if that's not the case with mine) Could it be possible that i solve this problem but i get some other panic (and so data loss)?
I think would be better to check with latest kernel as it is probably an upstream issue that is more probably fixed in 2.6.39
(In reply to comment #7) > I think would be better to check with latest kernel as it is probably an > upstream issue that is more probably fixed in 2.6.39 I'll try in a few hours.
Just tried the 2.6.39-gentoo-r1, same problem. I've also understood that it "panic" the second time i plug the device; so i think it's an address (irq) problem, probably it try to assign the same address again or something similar. (Sorry, i'm a little bit ignorant about that think...)
Then, provide "dmesg" output and kernel config for 2.6.39 kernel, thanks
Can't provide dmesg because of the panic... You can look at the attached jpeg. The .config it's the same as the 2.6.38
Fixed enabling kernel's CONFIG_USB_SUSPEND.