linux-kernel - Re: [PATCH] PCI/portdrv: Allow probing even without child services

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: <aYqHFGKDjVJ_mOtp@google.com>
Date: Mon, 9 Feb 2026 17:17:08 -0800
From: Brian Norris <briannorris@...omium.org>
To: Lukas Wunner <lukas@...ner.de>
Cc: Bjorn Helgaas <bhelgaas@...gle.com>, linux-pci@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Manivannan Sadhasivam <mani@...nel.org>
Subject: Re: [PATCH] PCI/portdrv: Allow probing even without child services

(Sorry for some delay. Other priorities take over sometimes...)

On Thu, Jan 15, 2026 at 09:44:40AM +0100, Lukas Wunner wrote:
> I note that pcie_portdrv_remove() calls pci_disable_device()
> unconditionally.  You may need an extra struct with an extra flag
> to remember whether pci_disable_device() needs to be called on remove.

Note that pci_enable_device() and pci_set_master() are separate
operations. With the proposed change, we may undo the latter, but not
the former. So remove() will still need to call pci_disable_device(),
which automatically handles clearing the master bit if needed, without
extra state tracking.

So I don't see the problem here.

At any rate, I've sent v2 that I think addresses everything. Feel free
to tell me I'm wrong there if needed :)

Brian

P.S. This whole concern is likely for naught if the port actually has a
device at the other end, since pci_enable_bridge() will set the master
bit anyway.