Messages in this thread |  | | Date | Wed, 27 Sep 2000 02:00:33 +1100 | From | Andrew Morton <> | Subject | Re: tty and ldisc module safety |
| |
Jeff Garzik wrote: > > On Wed, 27 Sep 2000, Andrew Morton wrote: > > > > --- linux-2.4.0-test9-pre7/include/linux/tty_driver.h Sat Aug 5 13:35:24 2000 > > +++ linux-akpm/include/linux/tty_driver.h Wed Sep 27 00:47:24 2000 > > @@ -117,6 +117,15 @@ > > > > #include <linux/fs.h> > > > > +#ifdef CONFIG_MODULES > > +#include <linux/module.h> > > +struct module; > > +#define SET_TTY_OWNER(driver) \ > > + do { (driver)->owner = THIS_MODULE; } while (0) > > +#define SET_LDISC_OWNER(ldisc) \ > > + do { (ldisc)->owner = THIS_MODULE; } while (0) > > +#endif > > Why ifdef MODULE?
So we can later do
#ifdef SET_TTY_OWNER
to determine in one hit whether we're compiling for a kernel which has this change and whether we're supporting modules. Much more reliable than LINUX_VERSION_CODE().
Those 4 bytes in the tty and ldisc structs may come in handy one day. I really dislike adding little bits of creeping bloat.
> And why the "struct module;"?
Stupidity?
> Simply including > linux/modules.h should take care of both cases... Like pci.h, modules.h > should work fine even when !MODULES...
Yep. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |