linux-kernel - Re: [PATCH v2] PCI: rockchip: Avoid accessing PCIe registers with clocks gated

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]

Date:   Thu, 1 Jul 2021 16:59:35 +0200
From:   Javier Martinez Canillas <javierm@...hat.com>
To:     Bjorn Helgaas <helgaas@...nel.org>
Cc:     linux-kernel@...r.kernel.org,
        Peter Robinson <pbrobinson@...il.com>,
        Shawn Lin <shawn.lin@...k-chips.com>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        Heiko Stuebner <heiko@...ech.de>,
        Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Rob Herring <robh@...nel.org>,
        linux-arm-kernel@...ts.infradead.org, linux-pci@...r.kernel.org,
        linux-rockchip@...ts.infradead.org,
        Michal Simek <michal.simek@...inx.com>,
        Jingoo Han <jingoohan1@...il.com>,
        Thierry Reding <thierry.reding@...il.com>,
        Jonathan Hunter <jonathanh@...dia.com>,
        linux-tegra@...r.kernel.org
Subject: Re: [PATCH v2] PCI: rockchip: Avoid accessing PCIe registers with
 clocks gated

On 7/1/21 3:59 PM, Bjorn Helgaas wrote:

[snip]

>> The IRQ handler is not called when unregistered, but it is called
>> when another handler for the shared IRQ is unregistered. In this
>> particular driver, both a "pcie-sys" and "pcie-client" handlers are
>> registered, then an error leads to "pcie-sys" being unregistered and
>> the handler for "pcie-client" being called.
> 
> Is this really true?  I think that would mean CONFIG_DEBUG_SHIRQ would
> not find this kind of bug unless we actually registered two or more
> handlers for the shared IRQ, but it's still a bug even only one
> handler is registered.
> 
> Looking at __free_irq() [1], my impression is that "action" is what
> we're removing and action->handler() is the IRQ handler we call when
> CONFIG_DEBUG_SHIRQ, so it doesn't look like it's calling the remaining
> handlers after removing one of them.
> 

Oh, you are completely right. I wrongly assumed that it was for the other
registered IRQ handlers but reading the source is clearly how you say it.

I now wonder why when debugging this I saw that the "pcie-client" handler
was called when "pcie-sys" was unregistered...

But anyways, you are correct and I'm OK with the text you shared.

Best regards,
-- 
Javier Martinez Canillas
Software Engineer
New Platform Technologies Enablement team
RHEL Engineering