[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: exec server and /dev/fd/N
From: |
Carl Fredrik Hammar |
Subject: |
Re: exec server and /dev/fd/N |
Date: |
Mon, 31 May 2010 18:27:29 +0200 |
User-agent: |
Mutt/1.5.20 (2009-06-14) |
Hi,
I have reviewed the patches and apart from formatting there were only
a couple of issues. Next iteration is hopefully the last.
On Thu, May 27, 2010 at 06:22:26PM +0200, Emilio Pozuelo Monfort wrote:
> On 25/05/10 21:10, Carl Fredrik Hammar wrote:
> > It would also be good if you always include a ChangeLog so I can catch
> > early errors there too, which hopefully leads to less round-trips.
>
> Writing ChangeLogs still takes me a long time, so I prefer to do it when
> the patches are close to finished so I don't need to rewrite them all
> the time, at least for big patches. If that's a big issue let me know
> and I'll write them all the time.
OK, no problem. I still think it would be a good idea though, because
after you have written it initially it is pretty easy to keep it up
to date. But you should include it in the next patches since they're
almost done.
> I've tested the following upgrade scenarios:
>
> * Upgrading only glibc (libc.so + libhurduser.so). System works fine.
> Reboot. System works fine.
>
> * Upgrading glibc (libc.so + libhurduser.so) first and then hurd (exec,
> ext2fs.static, lib{disks,triv,net}fs.so). System works fine. Reboot.
> System works fine.
>
> * Upgrading only hurd (exec, ext2fs.static, lib{disks,triv,net}fs.so).
> System works fine. Reboot. System doesn't boot. The problem is that
> the exec server dies because it needs the file_exec_file_name symbol,
> which is in libhurduser.so, lib*fs.so would have exec_exec_file_name
> unresolvable too... So it's not possible to upgrade just Hurd (which
> is fine IMHO, you need a new glibc to build the new Hurd anyway) unless
> we get into weird tricks OR we move the client side RPCs from
> libhurduser.so to a library built from Hurd.
Sounds good.
> Regarding version.h, I've bumped HURD_INTERFACE_VERSION in 0001 for
> exec_exec_file_name, but should it be bumped in 0002 too?
This sounds good to me but if someone else objects you should change it.
Regards,
Fredrik
- [PATCH 1/3] Add a new exec_exec_file_name RPC, (continued)
- [PATCH] Use the new __hurd_exec_file_name RPC, Emilio Pozuelo Monfort, 2010/05/27
- Re: [PATCH] Use the new __hurd_exec_file_name RPC, Carl Fredrik Hammar, 2010/05/31
- Re: [PATCH] Use the new __hurd_exec_file_name RPC, olafBuddenhagen, 2010/05/31
- [PATCH 3/3] Use the new _hurd_exec_file_name function, Emilio Pozuelo Monfort, 2010/05/27
- Re: [PATCH 3/3] Use the new _hurd_exec_file_name function, Carl Fredrik Hammar, 2010/05/31
- [PATCH 2/3] Add a file_exec_file_name RPC, Emilio Pozuelo Monfort, 2010/05/27
- Re: [PATCH 2/3] Add a file_exec_file_name RPC, Carl Fredrik Hammar, 2010/05/31
- Re: exec server and /dev/fd/N,
Carl Fredrik Hammar <=
- Re: exec server and /dev/fd/N, Carl Fredrik Hammar, 2010/05/31
Re: exec server and /dev/fd/N, olafBuddenhagen, 2010/05/26