Re: glibc developers refuse to support user land fake FS syscalls

Theodore Y. Ts'o (tytso@mit.edu)
Thu, 1 Jul 1999 20:54:38 -0400


From: Ulrich Drepper <drepper@cygnus.com>
Date: 01 Jul 1999 14:19:28 -0700

> Ulrich, this may be a problem for us, because there are some aspects of
> FS functionality that are easier/better implemented in user space.

You can do this, if you want. It would require replacing the syscalls
*in* the libc, and not replacing them from the outside. This is what
happens so far.

But this also will not work in all cases: old, statically linked
applications will not be changed, applications might make the syscalls
on their own without the help of the libc.

Therefore anything like this is error-prone.

If you want to handle something in userland then write code which is
called by the kernel in an upcall. This is the only reliable
solution. It might not be the fastest, though.

The other option is to do it as a global LD_PRELOAD, as I mentioned
earlier.

- Ted

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/