Andreas Jaeger wrote:
> Jamie> There just seems to be a hole at the moment: Glibc exports d_type info
> Jamie> from readdir but always stores DT_UNKNOWN in it.
>
> And that's why Ulrich Drepper likes to see a new syscall. If we've
> got a new syscall, we can easiliy store DT_UNKNOWN when using the old
> one - and using the real value from the kernel with the new syscall.
>
> For kernels without your changes, we need to set d_type to DT_UNKNOWN,
> otherwise applications will break.
Please read my response to Ulrich. The patch results in correct
behaviour and is binary compatible in all combinations, even with libc4
and libc5, without a new syscall.
A one line change to Glibc is sufficient. New Glibcs work correctly
with older kernels, and old Glibcs work correctly with newer kernels.
Applications always see either DT_UNKNOWN or, with up to date Glibc and
kernel, the correct type.
There are other reasons why another syscall may be desirable, but the
reason you give is not one of them.
-- Jamie
-
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/
This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:22 EST