[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: hurd does NOT need /hurd



On Tue, May 21, 2002 at 11:29:24PM +0200, Lars Weber wrote:
> All this talk about reasons for using `/hurd' got me wondering: Do there
> exist potential problems when a translator that translates a certain
> directory is itself located somewhere inside that directory?

No, because if you exec a file, you "open" it (what the Hurd equivalent of
the open it), and only pass the descriptor to the exec server (actually the
Hurd equivalent of a descriptor).

So, you keep an active reference to the file, and always talk to the
"underlying" translator.

There are issues with executing script.  I think for scripts it is the
normal case that they are reopened by name, but if that is not possible,
they are reopened by giving the interpreter the name /dev/fd/NR
 
> This might not be the case, but if it is I think this would be reason
> enough to use a separate directory.

/hurd is provided by a translator as well, and there can be translators on
/ or /hurd, so that doesn't save you.  If you start to talk about the
details it is quickly getting pretty weird :), but for example the
bootstrap is done by GRUB, which can read filesystems itself, so it reads
/hurd/ext2fs from the disk directly, and loads that into memory, in addition
to the microkernel and the run time linker.

Thanks,
Marcus


-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de


-- 
To UNSUBSCRIBE, email to debian-hurd-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org



Reply to: