I'm not sure if this is a usb-serial issue or a tty issue; I'm
starting here, since it's usb-serial's warning.
drivers/usb/serial/usb-serial.c is generating a warning in
serial_write(): port->port.count is 0:
<4>WARNING: at drivers/usb/serial/usb-serial.c:400
serial_write+0x5c/0x7b [usbserial]()
<4>Hardware name: G31M-ES2L
<4>Modules linked in: r8169 atl1c ppp_async ppp_generic slhc option
crc_ccitt rtc ftdi_sio usbserial usbhid nvidia(P) i2c_core evdev
ohci_hcd uhci_hcd agpgart forcedeth e100 snd_hda_codec_realtek
snd_hda_intel snd_hda_codec snd_intel8x0 snd_ac97_codec ac97_bus
snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd snd_page_alloc ub
usbcore
<4>Pid: 5, comm: events/0 Tainted: P W 2.6.31.6 #18
<4>Call Trace:
<4> [<c101855f>] ? warn_slowpath_common+0x5e/0x71
<4> [<c10185a7>] ? warn_slowpath_null+0xa/0x10
<4> [<e1dcf601>] ? serial_write+0x5c/0x7b [usbserial]
<4> [<c117a917>] ? tty_put_char+0x23/0x26
<4> [<c117b1a0>] ? do_output_char+0x18c/0x195
<4> [<c117b4e7>] ? process_echoes+0x1d6/0x265
<4> [<c117c209>] ? n_tty_receive_buf+0x69e/0x876
<4> [<c1012a12>] ? check_preempt_wakeup+0xd5/0xe6
<4> [<c1012401>] ? update_curr+0x65/0xb3
<4> [<c10149e4>] ? finish_task_switch+0x37/0x69
<4> [<c117ef55>] ? flush_to_ldisc+0xca/0x15f
<4> [<c117ee8b>] ? flush_to_ldisc+0x0/0x15f
<4> [<c10244ed>] ? run_workqueue+0x5e/0xf4
<4> [<c102460b>] ? worker_thread+0x88/0x94
<4> [<c1027210>] ? autoremove_wake_function+0x0/0x2d
<4> [<c126403a>] ? schedule+0x112/0x2f9
<4> [<c1027210>] ? autoremove_wake_function+0x0/0x2d
<4> [<c1024583>] ? worker_thread+0x0/0x94
<4> [<c1026d72>] ? kthread+0x6b/0x70
<4> [<c1026d07>] ? kthread+0x0/0x70
<4> [<c1002f93>] ? kernel_thread_helper+0x7/0x10
<4>---[ end trace f6ebfac90a49d5e1 ]---
This happens when /dev/ttyUSB0 is closed. The device continues
sending data indefinitely.
n_tty_receive_buf (drivers/char/n_tty.c) looks like it's at:
if (c == '\n') {
if (tty->read_cnt >= N_TTY_BUF_SIZE) {
if (L_ECHO(tty))
process_output('\a', tty);
return;
}
if (L_ECHO(tty) || L_ECHONL(tty)) {
echo_char_raw('\n', tty);
process_echoes(tty);
}
goto handle_newline;
}
under process_echoes. I'm not even sure why it's here, since ttyUSB0
is -icanon.
This doesn't reproduce reliably; when I reloaded usbserial.ko to check
something, it went away. It's not a pressing problem (it doesn't seem
to be breaking anything), though if it's actually ending up in icanon
it'd cause trouble.
--
Glenn Maynard
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
