Re: [PATCH] ttyprintk: Add TTY hangup callback.

[Samo Pogačnik]

Dne 23.04.2021 (pet) ob 19:12 +0900 je Tetsuo Handa napisal(a):
> On 2021/04/23 18:55, Samo Pogačnik wrote:
> > > > Using the supplied test code, i've tested the patch on my desktop
> > > > running
> > > > the
> > > > 5.4 kernel. After applying the patch, the kernel warnings like
> > > > "ttyprintk:
> > > > tty_port_close_start: tty->count = 1 port count = 11" do not appear any
> > > > more,
> > > > when the test code is run.
> > > > I think the patch is ok.
> > > 
> > > I wonder if the buffer shouldn't be flushed in hangup too? Or better, 
> > > the flush moved from tty_ops->close to tty_port->ops->shutdown?
> > > 
> > > thanks,
> > 
> > Good point. I tried the following additional change, which seems to do the
> > trick. What do you think?
> > 
> 
> Shouldn't the tpk_printk buffer be per a "struct file" (i.e. allocated upon
> open() and released upon close() in order to allow multiple users) ?

Final destination of the ttyprintk is printk(), which is a single destination.
The tpk_printk buffer is a common representation of what is yet to be printk-ed
depending on the formatting conditions within the buffer.

At any point the tpk_buffer is potentially multiplexed/interleaved by parts of
required output of any concurrent user, as buffs are being delivered by the
scheduled writes.

As per user buffers look promising with output formatting, the FDs passing
between tasks lead to the same single buffer (Greg already mentioned that). The
other thing which is important to me is the console redirection to ttyprintk,
which automatically brings all concurrent console users to the single open of
ttyprintk.

best regards, Samo