Messages in this thread | | | Date | Wed, 6 Mar 1996 16:21:24 -0500 | From | "Theodore Ts'o" <> | Subject | Re: struct stat - was: Re: Divergences ... |
| |
Date: Wed, 6 Mar 1996 03:21:08 +0100 From: Andries.Brouwer@cwi.nl
: You can't avoid it. How do you propose to make sure, then, that libc : and the kernel have the same notion of what struct stat is? (Just for : example.)
The idea is not that the libc include files never change. The idea is that they change once in three months, namely when one installs a new libc version. Not when one does a new "make config" in /usr/src/linux.
Well, if a kernel supports a new ioctl, we shouldn't have to wait for a new libc before we can take advantage of the ioctl. In the code I write, I very often will play games like:
#ifdef EXT2_IOCTL do_ioctl_code().... #else do_workaround()... #endif
If we tie the include files to the libc, then it makes it much harder for people to take advantage of some kernel features.
That's why we have some include files belonging to the kernel, and some include files distributed with libc. Yes, this means that we need to have coordination between kernel developers and libc developers. But this is always necessary! Unlike the BSD camp, we actually decouple the two, which I think is a feature. It has in the past led to faster development processes, IMO.
- Ted
| |