Re: [hurd-amd64] ibus test failures

[Diego Nieto Cid]

On Tue, Dec 31, 2024 at 02:52:53PM +0100, Samuel Thibault wrote:
> Samuel Thibault, le mar. 31 déc. 2024 14:36:02 +0100, a ecrit:
> > > I'm not sure how to fix it. Should the expectations of GLib be updated
> > > to be aligned? Like below:
> > > 
> > > - if (size != G_CREDENTIALS_NATIVE_SIZE)
> > > + if (size != CMSG_ALIGN(G_CREDENTIALS_NATIVE_SIZE))
> > > 
> > > Or should the macros be updated to use a 4 bytes size_t equivalent?
> > 
> > I'd rather say it's our socket layer which is aligning cmsg_len while
> > it shouldn't, since it's CMSG_NXTHDR that aligns it to get to the next
> > header.
> 
> Or perhaps it's glib itself that does the rounding, see
> 
> ./gio/gsocket.c:        cmsg->cmsg_len = CMSG_LEN 
> (g_socket_control_message_get_size (_message->control_messages[i]));
> 

This is OK, according to `man cmsg`

> Adding CMSG_ALIGN to the glib size check seems odd to be because in the
> linux case G_CREDENTIALS_NATIVE_SIZE is sizeof (struct ucred), which is
> 12bytes, so not 8byte-round and rounding would break the check on Linux.
> 
> You should probably check precisely the difference between Linux and
> Hurd on these alignment questions at the various stages.
> 

Ok, I'll have a look in both OSes behaviour.