linux-kernel - Re: [PATCH v11 09/14] irqchip/riscv-imsic: Add support for PCI MSI irqdomain

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: <87bkchfbnu.fsf@all.your.base.are.belong.to.us>
Date:   Sun, 29 Oct 2023 20:53:25 +0100
From:   Björn Töpel <bjorn@...nel.org>
To:     Thomas Gleixner <tglx@...utronix.de>,
        Anup Patel <apatel@...tanamicro.com>
Cc:     Palmer Dabbelt <palmer@...belt.com>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Frank Rowand <frowand.list@...il.com>,
        Conor Dooley <conor+dt@...nel.org>,
        Marc Zyngier <maz@...nel.org>,
        Atish Patra <atishp@...shpatra.org>,
        Andrew Jones <ajones@...tanamicro.com>,
        Sunil V L <sunilvl@...tanamicro.com>,
        Saravana Kannan <saravanak@...gle.com>,
        Anup Patel <anup@...infault.org>,
        linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
        devicetree@...r.kernel.org
Subject: Re: [PATCH v11 09/14] irqchip/riscv-imsic: Add support for PCI MSI
 irqdomain

Thomas Gleixner <tglx@...utronix.de> writes:

> On Wed, Oct 25 2023 at 10:55, Björn Töpel wrote:
>>> Now for IMSIC-PCI domain, the PCI framework expects the
>>> pci_msi_mask/unmask_irq() functions to be called but if
>>> we directly point pci_msi_mask/unmask_irq() in the IMSIC-PCI
>>> irqchip then IMSIC-BASE (parent domain) irq_mask/umask
>>> won't be called hence the IRQ won't be masked/unmask.
>>> Due to this, we call both pci_msi_mask/unmask_irq() and
>>> irq_chip_mask/unmask_parent() for IMSIC-PCI domain.
>>
>> Ok. I wont dig more into it for now! If the interrupt is disabled at
>> PCI, it seems a bit overkill to *also* mask it at the IMSIC level...
>
> Only _if_ the device provides MSI masking, but that extra mask/unmask is
> not the end of the world.

Yikes -- so MSI masking is optional. Ick. :-( Thanks for the excellent
MSI vs MSI-X post in the other thread, BTW. Great stuff!