linux-kernel - RE: [PATCH v3] ACPI: AGDI: Add interrupt signaling mode support

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:
 <TY3PR01MB9983E784E31A612B9BA23255D51BA@TY3PR01MB9983.jpnprd01.prod.outlook.com>
Date: Mon, 29 Sep 2025 07:32:12 +0000
From: "Kazuhiro Abe (Fujitsu)" <fj1078ii@...itsu.com>
To: 'Hanjun Guo' <guohanjun@...wei.com>, 'Ilkka Koskinen'
	<ilkka@...amperecomputing.com>, 'Sudeep Holla' <sudeep.holla@....com>
CC: 'Lorenzo Pieralisi' <lpieralisi@...nel.org>, "'Rafael J. Wysocki'"
	<rafael@...nel.org>, 'Len Brown' <lenb@...nel.org>,
	"'linux-acpi@...r.kernel.org'" <linux-acpi@...r.kernel.org>,
	"'linux-arm-kernel@...ts.infradead.org'"
	<linux-arm-kernel@...ts.infradead.org>, "'linux-kernel@...r.kernel.org'"
	<linux-kernel@...r.kernel.org>
Subject: RE: [PATCH v3] ACPI: AGDI: Add interrupt signaling mode support

Hi Hanjun,

> > Hi Hanjun & Sudeep
> >
> >> Hi Ilkka
> >>
> >>> Hi Kazuhiro,
> >>>
> >>> On Fri, 5 Sep 2025, Kazuhiro Abe wrote:
> >>>> AGDI has two types of signaling modes: SDEI and interrupt.
> >>>> Currently, the AGDI driver only supports SDEI.
> >>>> Therefore, add support for interrupt signaling mode The interrupt
> >>>> vector is retrieved from the AGDI table, and call panic function
> >>>> when an interrupt occurs.
> >>>>
> >>>> Signed-off-by: Kazuhiro Abe <fj1078ii@...jp.fujitsu.com>
> >>>
> >>>
> >>> Looks good to me.
> >>>
> >>> Reviewed-by: Ilkka Koskinen <ilkka@...amperecomputing.com>
> >>>
> >>
> >> Thanks for your review.
> >>
> >> Best Regards,
> >> Kazuhiro Abe
> >>
> >>>
> >>> Hanjun & Sudeep, what's your thought on enabling the use of regular
> >>> interrupts here? I do agree the spec talks about non-maskable ones
> >>> and to my understanding that's what the idea was indeed.
> >
> > Do you have any thoughts on this matter?
> 
> For the real use case, if the system is in failure state, for example, the system is
> panic, the CPU will not handle regular interrupts, so what's the use case do you
> have to use regular interrupt?

This driver is designed to operate with NMI enabled.

However, on current ARM64 platforms, NMI functionality may not be active unless
"pseudo NMI" is explicitly specified via a kernel command-line parameter.
In such scenarios, we've included regular interrupt handling as a last resort,
anticipating rare cases where some cores might still be able to receive regular
interrupts even if other cores are unresponsive.

To reiterate, the underlying assumption is that NMI(now pseudo NMI) is enabled for use.


Best Regards,
Kazuhiro Abe

> 
> Thanks
> Hanjun