On Mon, 2 Nov 2009, Nicolas Pitre wrote: > With this folded in, the card does work with the full series applied. > However the kernel is now crashing when the card is pulled out while > some process is reading from the device. This used to behave well > before. I don't have time to investigate that one right now though. OK, here more data points: My test is to insert GPS card and do a 'cat /dev/ttySDIO0' and watch NMEA data scrolling by. With "sdio_uart: Move the open lock" applied, I can pull the card out and the cat process terminates. However inserting the card back and repeating the test again always gives me: cat: /dev/ttySDIO0: Input/output error until I reboot. Same thing happens if I terminates cat with ^C and start it again. With "sdio_uart: Switch to the open/close helpers" applied, things got even worse. Interrupting cat with ^C still gives the same result as above. But yanking the card out when cat is still running doesn't terminate cat unless I ^C it at which point the kernel crashes with a BUG(). The backtrace is: [<c0029c88>] (__bug+0x18/0x24) from [<c02665d0>] (sdio_writeb+0x1c/0x4c) [<c02665d0>] (sdio_writeb+0x1c/0x4c) from [<c01b4098>] (tty_port_lower_dtr_rts+0x1c/0x20) [<c01b4098>] (tty_port_lower_dtr_rts+0x1c/0x20) from [<c01b412c>] (tty_port_close_end+0x2c/0x130) [<c01b412c>] (tty_port_close_end+0x2c/0x130) from [<c01b49a8>] (tty_port_close+0x2c/0x3c) [<c01b49a8>] (tty_port_close+0x2c/0x3c) from [<c01ae0c4>] (tty_release_dev+0x17c/0x478) [<c01ae0c4>] (tty_release_dev+0x17c/0x478) from [<c01ae3e8>] (tty_release+0x28/0x4c) [<c01ae3e8>] (tty_release+0x28/0x4c) from [<c00b5bf0>] (__fput+0x118/0x210) [<c00b5bf0>] (__fput+0x118/0x210) from [<c00b28b4>] (filp_close+0x70/0x7c) [<c00b28b4>] (filp_close+0x70/0x7c) from [<c003d0e0>] (put_files_struct+0x80/0xd4) [<c003d0e0>] (put_files_struct+0x80/0xd4) from [<c003eac0>] (do_exit+0x220/0x714) [<c003eac0>] (do_exit+0x220/0x714) from [<c003f074>] (do_group_exit+0xc0/0xf4) [<c003f074>] (do_group_exit+0xc0/0xf4) from [<c004d8d0>] (get_signal_to_deliver+0x3c8/0x428) [...] (impressive nesting culled) Only after backing out those 2 patches as wellas "sdio_uart: refcount the tty objects" is the correct behavior restored. Nicolas [<c004d8d0>] (get_signal_to_deliver+0x3c8/0x428) from [<c0028b34>] (do_notify_resume+0x5c/0x5d0) [<c0028b34>] (do_notify_resume+0x5c/0x5d0) from [<c0026a88>] (work_pending+0x1c/0x20) > > > Nicolas > -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
