Im not completely sure but aparently the new kms radeon driver in staging is making the usbserial (in my case a 3g modem) to crash upon calling the device. this behaviour is reproducible every time in 2.6.31 rc3, it doesnt happen in rc2. Jul 22 16:10:31 kerberos kernel: [ 344.869682] ------------[ cut here ]------------ Jul 22 16:10:31 kerberos kernel: [ 344.869748] WARNING: at drivers/usb/serial/usb-serial.c:436 serial_ioctl+0x95/0xa0 [usbserial]() Jul 22 16:10:31 kerberos kernel: [ 344.869756] Hardware name: Aspire 5100 Jul 22 16:10:31 kerberos kernel: [ 344.869761] Modules linked in: i2c_dev cbc aes_i586 aes_generic ramzswap xvmalloc lzo_decompress lzo_compre ss binfmt_misc ppdev vboxdrv lp parport snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm arc4 snd_ seq_dummy ecb snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq ath5k snd_timer snd_seq_device mac80211 pcmcia ath snd joydev sou ndcore yenta_socket rsrc_nonstatic pcmcia_core i2c_piix4 sdhci_pci sdhci k8temp cfg80211 ati_agp psmouse serio_raw pcspkr snd_page_alloc video output shpchp acer_wmi led_class reiserfs 8139too 8139cp mii sg sd_mod usbhid usb_storage option usbserial fbcon tileblit font bitblit softcurs or radeon i2c_algo_bit ttm drm agpgart Jul 22 16:10:31 kerberos kernel: [ 344.869914] Pid: 3687, comm: NetworkManager Not tainted 2.6.31-rc3-custom #1 Jul 22 16:10:31 kerberos kernel: [ 344.869921] Call Trace: Jul 22 16:10:31 kerberos kernel: [ 344.869943] [<c055cb90>] ? printk+0x18/0x20 Jul 22 16:10:31 kerberos kernel: [ 344.869969] [<f82e8705>] ? serial_ioctl+0x95/0xa0 [usbserial] Jul 22 16:10:31 kerberos kernel: [ 344.869984] [<c014492c>] warn_slowpath_common+0x6c/0xc0 Jul 22 16:10:31 kerberos kernel: [ 344.870008] [<f82e8705>] ? serial_ioctl+0x95/0xa0 [usbserial] Jul 22 16:10:31 kerberos kernel: [ 344.870018] [<c0144995>] warn_slowpath_null+0x15/0x20 Jul 22 16:10:31 kerberos kernel: [ 344.870041] [<f82e8705>] serial_ioctl+0x95/0xa0 [usbserial] Jul 22 16:10:31 kerberos kernel: [ 344.870052] [<c012f4e0>] ? __wake_up_common+0x40/0x70 Jul 22 16:10:31 kerberos kernel: [ 344.870074] [<f82e8670>] ? serial_ioctl+0x0/0xa0 [usbserial] Jul 22 16:10:31 kerberos kernel: [ 344.870085] [<c0377f12>] tty_ioctl+0x82/0x890 Jul 22 16:10:31 kerberos kernel: [ 344.870155] [<f83d9362>] ? radeon_fence_process+0x52/0x60 [radeon] Jul 22 16:10:31 kerberos kernel: [ 344.870164] [<c0377e90>] ? tty_ioctl+0x0/0x890 Jul 22 16:10:31 kerberos kernel: [ 344.870176] [<c01f4e48>] vfs_ioctl+0x28/0x80 Jul 22 16:10:31 kerberos kernel: [ 344.870184] [<c01f4fda>] do_vfs_ioctl+0x6a/0x5e0 Jul 22 16:10:31 kerberos kernel: [ 344.870197] [<c0470386>] ? sys_send+0x36/0x40 Jul 22 16:10:31 kerberos kernel: [ 344.870207] [<c04709ad>] ? sys_socketcall+0x1cd/0x2a0 Jul 22 16:10:31 kerberos kernel: [ 344.870217] [<c01f55b3>] sys_ioctl+0x63/0x70 Jul 22 16:10:31 kerberos kernel: [ 344.870229] [<c0103c94>] sysenter_do_call+0x12/0x28 Jul 22 16:10:31 kerberos kernel: [ 344.870237] ---[ end trace 352de30743b9a960 ]--- could the radeon devs take a look and see if him right,or this is in fact a usbserial bug, in that case ill open a lkms bug report. best regards
Please try with lastest linux tree and report back (fixes wants in since r3)
yes, its fixed, thanks
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.