From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marc Zyngier Subject: Re: [PATCH 1/4] Irqchip: Ingenic: Change interrupt handling form cascade to chained_irq. Date: Sun, 27 Jan 2019 10:21:40 +0000 Message-ID: <86o982wgcr.wl-marc.zyngier@arm.com> References: <1548517123-60058-1-git-send-email-zhouyanjie@zoho.com> <1548517123-60058-2-git-send-email-zhouyanjie@zoho.com> Mime-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue") Content-Type: text/plain; charset=US-ASCII Return-path: In-Reply-To: <1548517123-60058-2-git-send-email-zhouyanjie@zoho.com> Sender: linux-kernel-owner@vger.kernel.org To: Zhou Yanjie Cc: linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org, devicetree@vger.kernel.org, robh+dt@kernel.org, paul.burton@mips.com, mark.rutland@arm.com, jason@lakedaemon.net, tglx@linutronix.de, syq@debian.org, jiaxun.yang@flygoat.com, 772753199@qq.com List-Id: devicetree@vger.kernel.org On Sat, 26 Jan 2019 15:38:40 +0000, Zhou Yanjie wrote: > > The interrupt handling method is changed from old-style cascade to > chained_irq which is more appropriate. Also, it can process the > corner situation that more than one irq is coming to a single > chip at the same time. > > Signed-off-by: Zhou Yanjie > --- > drivers/irqchip/irq-ingenic.c | 49 ++++++++++++++++++++++--------------------- > 1 file changed, 25 insertions(+), 24 deletions(-) > > diff --git a/drivers/irqchip/irq-ingenic.c b/drivers/irqchip/irq-ingenic.c > index 2ff0898..2713ec4 100644 > --- a/drivers/irqchip/irq-ingenic.c > +++ b/drivers/irqchip/irq-ingenic.c > @@ -1,16 +1,7 @@ > +// SPDX-License-Identifier: GPL-2.0 > /* > * Copyright (C) 2009-2010, Lars-Peter Clausen > - * JZ4740 platform IRQ support > - * > - * This program is free software; you can redistribute it and/or modify it > - * under the terms of the GNU General Public License as published by the > - * Free Software Foundation; either version 2 of the License, or (at your > - * option) any later version. > - * > - * You should have received a copy of the GNU General Public License along > - * with this program; if not, write to the Free Software Foundation, Inc., > - * 675 Mass Ave, Cambridge, MA 02139, USA. > - * > + * Ingenic XBurst platform IRQ support > */ > > #include > @@ -19,6 +10,7 @@ > #include > #include > #include > +#include > #include > #include > #include > @@ -41,22 +33,35 @@ struct ingenic_intc_data { > #define JZ_REG_INTC_PENDING 0x10 > #define CHIP_SIZE 0x20 > > -static irqreturn_t intc_cascade(int irq, void *data) > +static void ingenic_chained_handle_irq(struct irq_desc *desc) > { > - struct ingenic_intc_data *intc = irq_get_handler_data(irq); > - uint32_t irq_reg; > + struct ingenic_intc_data *intc = irq_desc_get_handler_data(desc); > + struct irq_chip *chip = irq_desc_get_chip(desc); > + bool have_irq = false; > + u32 pending; > unsigned i; > > + chained_irq_enter(chip, desc); > for (i = 0; i < intc->num_chips; i++) { > - irq_reg = readl(intc->base + (i * CHIP_SIZE) + > + pending = readl(intc->base + (i * CHIP_SIZE) + > JZ_REG_INTC_PENDING); > - if (!irq_reg) > + if (!pending) > continue; > > - generic_handle_irq(__fls(irq_reg) + (i * 32) + JZ4740_IRQ_BASE); > + have_irq = true; > + while (pending) { > + int bit = __ffs(pending); So 'bit' is the least significant bit in the pending word, > + > + generic_handle_irq(__fls(pending) + (i * 32) + and here you handle the *most significant* bit, > + JZ4740_IRQ_BASE); > + pending &= ~BIT(bit); yet it is the least significant bit that you clear. I am tempted to say that you have never tested this code with more than a single interrupt. Thanks, M. -- Jazz is not dead, it just smell funny.