linux-kernel - Re: [PATCH v3 2/2] s390/pci: Migrate s390 IRQ logic to IRQ domain API

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] [day] [month] [year] [list]

Message-ID: <871plum3x0.ffs@tglx>
Date: Wed, 19 Nov 2025 16:35:07 +0100
From: Thomas Gleixner <tglx@...utronix.de>
To: Tobias Schumacher <ts@...p.linux.ibm.com>, Farhan Ali <alifm@...ux.ibm.com>
Cc: Tobias Schumacher <ts@...ux.ibm.com>, Heiko Carstens
 <hca@...ux.ibm.com>, Vasily Gorbik <gor@...ux.ibm.com>, Alexander Gordeev
 <agordeev@...ux.ibm.com>, Christian Borntraeger
 <borntraeger@...ux.ibm.com>, Sven Schnelle <svens@...ux.ibm.com>, Niklas
 Schnelle <schnelle@...ux.ibm.com>, Gerald Schaefer
 <gerald.schaefer@...ux.ibm.com>, Gerd Bayer <gbayer@...ux.ibm.com>, Halil
 Pasic <pasic@...ux.ibm.com>, Matthew Rosato <mjrosato@...ux.ibm.com>,
 linux-kernel@...r.kernel.org, linux-s390@...r.kernel.org
Subject: Re: [PATCH v3 2/2] s390/pci: Migrate s390 IRQ logic to IRQ domain API

On Wed, Nov 19 2025 at 11:12, Tobias Schumacher wrote:
> Am 2025-11-18 22:49, schrieb Farhan Ali:
>> Also thinking it out loud, is it this going to be unique if we have
>> multiple IRQ (if nr_irqs in zpci_msi_domain_alloc() is > 1) per MSI
>> descriptor, unless I missed something?
>
> AFAIU also for MSI there is one MSI descriptor per IRQ (allocated in 
> msi_domain_alloc_simple_msi_descs()).

Correct because MSI only stores the MSI message for the index=0. If
there are multiple interrupts in the device then the MSI controller adds
the index of the interrupt to be raised in the lower bits of msi_msg::data

> The MSI index is incremented with each descriptor. We could iterate
> these descriptors, but since the index is incremented with each desc
> alloc() simply implements a for loop that increments the hwirq number
> for each irq.

That's correct because the MSI controller in the PCI device requires
that the hardware interrupt numbers are consecutive due to the above.

Thanks,

        tglx