On Thu, 25 Apr 2019 11:20:03 +0200
Petr Mladek wrote:
> >
> > Petr, Steven, would you prefer to have it as two separate patches - one
> > removes the comment; the other one moves pr_info("console enabled") - or
> > as one patch? It's sort of trivial enough to be in just one patch, but
> > I also
On (04/25/19 16:56), Sergey Senozhatsky wrote:
> > Great idea!
> >
> > It would deserve a separate patch that moves the pr_info()
> > and removes the invalid comment.
> >
> > Actually, the pr_info() would deserve a comment explaining
> > why it should be called before console_unlock().
>
> Good.
On Thu 2019-04-25 15:43:21, Sergey Senozhatsky wrote:
> On (04/25/19 12:52), Sergey Senozhatsky wrote:
> > > Could we remove it in this patch? It touches it indirectly anyway.
> >
> > Sure we can.
> >
> > We also can take extra care of pr_info("%sconsole [%s%d] enabled\n".
> > Right now we do
> >
On (04/25/19 09:50), Petr Mladek wrote:
> > Sure we can.
> >
> > We also can take extra care of pr_info("%sconsole [%s%d] enabled\n".
> > Right now we do
> >
> > ...
> > console_unlock();
> > console_sysfs_notify();
> >
> > pr_info("%sconsole [%s%d] enabled\n",
> >
> >
> >
On Thu 2019-04-25 12:52:33, Sergey Senozhatsky wrote:
> On (04/24/19 16:49), Petr Mladek wrote:
> > > + if (bcon && (newcon->flags & (CON_CONSDEV|CON_BOOT)) == CON_CONSDEV) {
> > > + console_lock();
> > > + /*
> > > + * We need to iterate through all boot consoles, to make
On (04/25/19 12:52), Sergey Senozhatsky wrote:
> > Could we remove it in this patch? It touches it indirectly anyway.
>
> Sure we can.
>
> We also can take extra care of pr_info("%sconsole [%s%d] enabled\n".
> Right now we do
>
> ...
> console_unlock();
> console_sysfs_notify()
On (04/24/19 16:49), Petr Mladek wrote:
> > + if (bcon && (newcon->flags & (CON_CONSDEV|CON_BOOT)) == CON_CONSDEV) {
> > + console_lock();
> > + /*
> > +* We need to iterate through all boot consoles, to make
> > * sure we print everything out, before
On Tue 2019-04-23 15:25:10, Sergey Senozhatsky wrote:
> The following pattern is not completely safe:
>
> for_each_console(bcon)
> if (bcon->flags & CON_BOOT)
> unregister_console(bcon);
>
> Because, in theory, console drivers list and console drivers
> can be modified
The following pattern is not completely safe:
for_each_console(bcon)
if (bcon->flags & CON_BOOT)
unregister_console(bcon);
Because, in theory, console drivers list and console drivers
can be modified concurrently from another CPU. Take console_sem
lock, which protects c
9 matches
Mail list logo
