linux-kernel - Re: genirq: Setting trigger mode 0 for irq 11 failed (txx9_irq_set

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <57DBA464.9010506@arm.com>
Date:   Fri, 16 Sep 2016 08:51:00 +0100
From:   Marc Zyngier <marc.zyngier@....com>
To:     Alban Browaeys <alban.browaeys@...il.com>,
        Geert Uytterhoeven <geert@...ux-m68k.org>,
        Atsushi Nemoto <anemo@....ocn.ne.jp>
Cc:     Linux MIPS Mailing List <linux-mips@...ux-mips.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Jon Hunter <jonathanh@...dia.com>
Subject: Re: genirq: Setting trigger mode 0 for irq 11 failed
 (txx9_irq_set_type+0x0/0xb8)

Hi Alban,

On 16/09/16 00:02, Alban Browaeys wrote:
> Le mercredi 14 septembre 2016 à 21:25 +0200, Geert Uytterhoeven a
> écrit :
>> JFYI, with v4.8-rc6 I'm seeing
>>
>>     genirq: Setting trigger mode 0 for irq 11 failed
>> (txx9_irq_set_type+0x0/0xb8)
>>
>> on rbtx4927. This did not happen with v4.8-rc3.
> 
> 
> txx9_irq_set_type receives a type IRQ_TYPE_NONE from the call to
> __irq_set_trigger added in:
> 1e12c4a939 ("genirq: Correctly configure the trigger on chained interrupts")
> 
> 
> This patch is a regression fix for :
> 
> Desc: irqdomain: Don't set type when mapping an IRQ breaks nexus7 gpio buttons
> Repo: 2016-07-30 https://marc.info/?l=linux-kernel&m=146985356305280&w=2
> 
> I am seeing this on arm odroid u2 devicetree :
> genirq: Setting trigger mode 0 for irq 16 failed (gic_set_type+0x0/0x64)

Passing IRQ_TYPE_NONE to a cascading interrupt is risky at best...
Can you point me to the various DTs and their failing interrupts?

Also, can you please give the following patch a go and let me know
if that fixes the issue (I'm interested in the potential warning here).

diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 6373890..8422779 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -820,6 +820,8 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle,
 	desc->name = name;
 
 	if (handle != handle_bad_irq && is_chained) {
+		unsigned int type = irqd_get_trigger_type(&desc->irq_data);
+
 		/*
 		 * We're about to start this interrupt immediately,
 		 * hence the need to set the trigger configuration.
@@ -828,8 +830,10 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle,
 		 * chained interrupt. Reset it immediately because we
 		 * do know better.
 		 */
-		__irq_set_trigger(desc, irqd_get_trigger_type(&desc->irq_data));
-		desc->handle_irq = handle;
+		if (!(WARN_ON(type == IRQ_TYPE_NONE))) {
+			__irq_set_trigger(desc, type);
+			desc->handle_irq = handle;
+		}
 
 		irq_settings_set_noprobe(desc);
 		irq_settings_set_norequest(desc);

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...