Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 22 Nov 2022 14:41:27 -0500
From:   "Gabriel L. Somlo" <gsomlo@gmail.com>
To:     Geert Uytterhoeven <geert@linux-m68k.org>
Cc:     Jiri Slaby <jirislaby@kernel.org>, linux-kernel@vger.kernel.org,
        linux-serial@vger.kernel.org, gregkh@linuxfoundation.org,
        kgugala@antmicro.com, mholenko@antmicro.com, joel@jms.id.au,
        david.abdurachmanov@gmail.com, florent@enjoy-digital.fr,
        ilpo.jarvinen@linux.intel.com
Subject: Re: [PATCH v5 12/14] serial: liteuart: add IRQ support for the RX
 path
Message-ID: <Y30l53r3SJCHiXjj@errol.ini.cmu.edu>
References: <20221118145512.509950-1-gsomlo@gmail.com>
 <20221118145512.509950-13-gsomlo@gmail.com>
 <abb7d87a-46c7-fe7f-f943-4287d15a537f@kernel.org>
 <Y3vIi2OF5t4IrCS1@errol.ini.cmu.edu>
 <8dd90e39-087e-984c-c289-dbfa766359a8@kernel.org>
 <CAMuHMdV3_YhF454J3ixRjOgWzJhL_Qr_XqxtSqxjF0qxGotFQw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAMuHMdV3_YhF454J3ixRjOgWzJhL_Qr_XqxtSqxjF0qxGotFQw@mail.gmail.com>
Precedence: bulk

On Tue, Nov 22, 2022 at 10:54:45AM +0100, Geert Uytterhoeven wrote:
> On Tue, Nov 22, 2022 at 8:44 AM Jiri Slaby <jirislaby@kernel.org> wrote:
> > On 21. 11. 22, 19:50, Gabriel L. Somlo wrote:
> > >>>    static void liteuart_timer(struct timer_list *t)
> > >>>    {
> > >>>     struct liteuart_port *uart = from_timer(uart, t, timer);
> > >>>     struct uart_port *port = &uart->port;
> > >>> -   liteuart_rx_chars(port);
> > >>> -
> > >>> +   liteuart_interrupt(0, port);
> > >>
> > >> Are you sure spin_lock() is safe from this path? I assume so, but have you
> > >> thought about it?
> > >
> > > I checked and at that point `in_serving_softirq()` is true.
> > >
> > > *However*, after studying spin_lock() and friends for a while, I'm
> > > not quite clear on whether that unequivocally translates
> > > to "yes, we're safe" :)
> >
> > Depends whether some hard irq context is grabbing the port lock too. If
> > it does, it will spin forever waiting for this soft irq to finish (never
> > happens as it was interrupted).
> >
> > > As such, I'm inclined to switch to `spin_lock_irqsave()` and
> > > `spin_unlock_irqrestore()` even in the interrupt handler, which is
> > > explicitly stated to be "safe from any context":
> > > https://www.kernel.org/doc/html/v4.15/kernel-hacking/locking.html#cheat-sheet-for-locking
> >
> >
> >
> > > The alternative could be to set `TIMER_IRQSAFE` in `timer_setup()`,
> > > but no other tty driver seems to be doing that, so I'd be a bit off
> > > the beaten path there... :)
> >
> > Ah, no.
> >
> > > Please do let me know what you think about this, particularly if you
> > > consider going the spin_lock_irqsave-everywhere-just-to-be-safe route
> > > overkill... :)
> >
> > If you are unsure about the other contexts, irqsave/restore is the way
> > to go. It can be lifted later, if someone investigates harder.
> 
> Inside the interrupt handler, plain spin_lock() should be OK.
> Inside the timer handler, I think you need to use spin_lock_irqsave(),
> as e.g. liteuart_set_termios() (and lots of code in serial_core.c)
> already uses spin_lock_irqsave().
> Besides, on non-SMP, spin_lock() doesn't do anything, so you need
> to rely on disabling the local interrupt.

Thanks Geert for the clarification! I could write two wrappers around
the actual code doing the interrupt handler work, one with spin_lock()
for the actual irq handler and another with spin_lock_irqsave() for
the timer codepath. But to keep things simple I'm inclined to simply
use spin_lock_irqsave() and add a comment saying it's because we need
it when polling and it's not actually harmful when using IRQ.

Thanks,
--Gabriel
 
> Gr{oetje,eeting}s,
> 
>                         Geert
> 
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
> 
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
>                                 -- Linus Torvalds