On Mon, Feb 24, 2025 at 2:23 AM Adam Simonelli <adamsimonelli@xxxxxxxxx> wrote: > On Sunday, February 23, 2025 4:10:10 PM EST Andy Shevchenko wrote: > > Sun, Feb 23, 2025 at 03:44:54PM -0500, adamsimonelli@xxxxxxxxx kirjoitti: > > > From: Adam Simonelli <adamsimonelli@xxxxxxxxx> > > > > > > When switching to a CONFIG_VT=n world, at least on x86 systems, > > > /dev/console becomes /dev/ttyS0. This can cause some undesired effects. > > > /dev/console's behavior is now tied to the physical /dev/ttyS0, which when > > > disconnected can cause isatty() to fail when /dev/ttyS0 is disconnected, > > > and users who upgrade to a theoretical vt-less kernel from their > > > distribution who have a device such as a science instrument connected to > > > their /dev/ttyS0 port will suddenly see it receive kernel log messages. > > > > > > When the new CONFIG_NULL_TTY_CONSOLE option is turned on, this will allow > > > the ttynull device to be leveraged as the default console. Distributions > > > that had CONFIG_VT turned on before will be able to leverage this option > > > to where /dev/console is still backed by a psuedo device, avoiding these > > > issues, without needing to enable the entire VT subsystem. > > > > This rings a bell of the following > > > > https://lore.kernel.org/all/20201111135450.11214-1-pmladek@xxxxxxxx/ > > https://lore.kernel.org/all/20210107164400.17904-1-pmladek@xxxxxxxx/ > > https://lore.kernel.org/all/20210108114847.23469-1-pmladek@xxxxxxxx/ > > > > I don't see any mention in the commit message about these, have you studied the > > cases? Will your change anyhow affect the described there? > > > I did see that sifting through commits, it looks kind of different though, as > that was to make ttynull more always on, and if I am understanding it correctly > it looks more of a last resort? I could be wrong, but I see it was attempted > and reverted because of conflicts on some hardware platforms? > > The scope with this new patch set is much different, as it has to be manually > enabled on top of CONFIG_NULL_TTY, rather than assuming CONFIG_NULL_TTY is > enabled, and adding it to the list of preferred consoles. It is more for > Desktop configs that are looking to disable CONFIG_VT, but still want > /dev/console to not be a physical device by default. Thank you for elaboration. Even if this may be a different case, I think it would be nice to have the PRINTK maintainers (who are also involved in the console code) blessing before going in. In case of a new version, please summarize above in the commit message. -- With Best Regards, Andy Shevchenko
