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:   Fri, 29 Apr 2022 18:51:44 +0100
From:   Daniel Thompson <daniel.thompson@linaro.org>
To:     Ard Biesheuvel <ardb@kernel.org>
Cc:     Marc Zyngier <maz@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] irqchip/exiu: Fix acknowledgment of edge triggered
 interrupts
Message-ID: <20220429175144.czymk5alqvqobzzf@maple.lan>
References: <20220427142647.1736658-1-daniel.thompson@linaro.org>
 <CAMj1kXEYO_HRwZ=Wz5aTWN8XZctemVv2op4B2=DbFEozuxTNVw@mail.gmail.com>
 <20220429132858.gaqhjrxblciacvki@maple.lan>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220429132858.gaqhjrxblciacvki@maple.lan>
Precedence: bulk

On Fri, Apr 29, 2022 at 02:28:58PM +0100, Daniel Thompson wrote:
> On Wed, Apr 27, 2022 at 05:29:33PM +0200, Ard Biesheuvel wrote:
> > On Wed, 27 Apr 2022 at 16:27, Daniel Thompson
> > <daniel.thompson@linaro.org> wrote:
> > >
> > > Currently the EXIU uses the fasteoi interrupt flow that is configured by
> > > it's parent (irq-gic-v3.c). With this flow the only chance to clear the
> > > interrupt request happens during .irq_eoi() and (obviously) this happens
> > > after the interrupt handler has run. EXIU requires edge triggered
> > > interrupts to be acked prior to interrupt handling. Without this we
> > > risk incorrect interrupt dismissal when a new interrupt is delivered
> > > after the handler reads and acknowledges the peripheral but before the
> > > irq_eoi() takes place.
> > >
> > > Fix this by clearing the interrupt request from .irq_ack() instead. This
> > > requires switching to the fasteoi-ack flow instead of the fasteoi flow.
> > > This approach is inspired by the nmi code found in irq-sun6i-r.c .
> > >
> > 
> > How are level triggered EXIU interrupts affected by this change?
> 
> Functionally they should not be affected simply because the controller
> does not care when (or even if) software writes to IREQCLR... and
> testing on SC2A11/Developerbox with a hacked gpio-keys driver does
> back this up.

If, and only if, you do the experiment properly... and this afternoon
the penny dropped and I fixed that.

In summary, level triggered interrupts require IREQCLR to happen at EOI
(to avoid spurious re-entry) whilst edge triggered interrupts require
IREQCLR to be acked before ISR is entered (to avoid spurious dismissal).

v2 is on its way...


Daniel.