On Mon, Feb 24, 2014 at 09:41:10AM -0500, Ray Strode wrote: > Hi, > > On Mon, Feb 24, 2014 at 3:38 AM, Dr. Werner Fink <[email protected]> wrote: > > IMHO this is a bug in plymouth as systemd does it correct in > Hey thanks for looking into this! > > It's sort of debatable whether or not it's a bug in plymouth, but I > think we can all agree it's not that useful of a debate to spend a lot > of time on. I'll just say one idea most of us can probably agree with > is: "When in doubt turn off the splash" which is what happened here. > It didn't see something it expected to see and so it turned off the > splash.
Indeed the users seems to be depend on the splash ;)
> > only checks for [...] tty0 but ignores the real active virtual console e.g.
> > tty1.
> Right, that's crux of the issue. The file used to just have tty0 and
> now resolves to the
> foreground console. (see the other thread for more specifics).
>
> > I've attached a simple patch to enhance plymouth's in the
> > same manner as systemd does it.
>
> Thanks for that. I actually already came up with a change that gets
> things working in a slightly different way for rawhide. I haven't
> decided if i want to push it upstream yet or not.
>
> Even, though, if we fix the code going forward, new kernels should
> continue to work with old plymouths, right? What I'm saying is, no
> matter where the "best" place to fix the problem in the stack is, we
> have to fix the issue in the kernel, because the kernel supports being
> swapped into older stacks without them being updated.
With a step further such a patch would allow to have both a splash
screen on the virtual consoles and boot messages on all other consoles.
It is not that unusal to use a serial console out there, is it :)
Beside this IMHO plymouth may add O_NOCTTY to all open() of the
devices combinded with setsid() after its fork() to become a daemon :)
The question rises if upstream of plymouth will read this?
Werner
--
"Having a smoking section in a restaurant is like having
a peeing section in a swimming pool." -- Edward Burr
pgpqGiJ5L5Nhc.pgp
Description: PGP signature
