Re: [PATCH 2/3] TTY: add support for tty_slave devices.
From: Pavel Machek <hidden>
Date: 2015-03-20 19:41:54
Also in:
lkml
Hi! (And yes, I now see dts examples, sorry for the noise.) Acked-by: Pavel Machek <redacted> Minor nits below.
quoted hunk ↗ jump to hunk
--- /dev/null +++ b/drivers/tty/slave/tty_slave_core.c@@ -0,0 +1,136 @@ +/* + * tty-slave-core - device bus for tty slaves
Filename actually uses underscores.
+ container_of(parent, struct tty_slave, dev); + tty->ops = &dev->ops; + } +} +EXPORT_SYMBOL(tty_slave_activate);
Not "_GPL"?
+postcore_initcall(tty_slave_init); +module_exit(tty_slave_exit);
Should it have MODULE_LICENSE tag?
+int tty_register_finalize(struct tty_driver *driver, struct device *dev)
+{
+ int retval;
+ bool cdev = false;
+ int index = dev->devt - MKDEV(driver->major,
+ driver->minor_start);
+ printk("REGISTER %d %d 0x%x %d\n", driver->major, driver->minor_start, dev->devt, index);That printk should probably be removed for merge?
+ if (!(driver->flags & TTY_DRIVER_DYNAMIC_ALLOC)) {
+ retval = tty_cdev_add(driver,
+ dev->devt,
+ index, 1);You can put this on one line.
quoted hunk ↗ jump to hunk
--- /dev/null +++ b/include/linux/tty_slave.h@@ -0,0 +1,26 @@ + +struct tty_slave { + struct device *tty_dev; + struct tty_driver *tty_drv; + struct tty_operations ops; + struct device dev; +};
Header files usually have #include guards, and some kind of comment on top. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html