The same as in "TTY: serial, remove BTM from wait_until_sent" we don't
need to take BTM in wait_until_sent of ami_serial. Exactly the same
as serial, ami_serial accesses some "info" members (xmit_fifo_size,
timeout), but their assignment on other places in the code is not
protected by BTM anyway.
So the BTM protects nothing here. This removal helps us to get rid of
tty_locked() and __big_tty_mutex_owner in the following patch. This
was suggested by Arnd.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Andreas Bombe <aeb@debian.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
{
struct async_struct * info = tty->driver_data;
unsigned long orig_jiffies, char_time;
- int tty_was_locked = tty_locked();
int lsr;
if (serial_paranoia_check(info, tty->name, "rs_wait_until_sent"))
orig_jiffies = jiffies;
- /*
- * tty_wait_until_sent is called from lots of places,
- * with or without the BTM.
- */
- if (!tty_was_locked)
- tty_lock();
/*
* Set the check interval to be 1/5 of the estimated time to
* send a single character, and make it at least 1. The check
break;
}
__set_current_state(TASK_RUNNING);
- if (!tty_was_locked)
- tty_unlock();
+
#ifdef SERIAL_DEBUG_RS_WAIT_UNTIL_SENT
printk("lsr = %d (jiff=%lu)...done\n", lsr, jiffies);
#endif