linux-kernel - Re: [PATCH v2 3/5] PCI/pwrctl: Ensure that the pwrctl drivers are probed before the PCI client drivers

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: <20241120201839.GA2338274@bhelgaas>
Date: Wed, 20 Nov 2024 14:18:39 -0600
From: Bjorn Helgaas <helgaas@...nel.org>
To: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
Cc: Bjorn Helgaas <bhelgaas@...gle.com>,
	Bartosz Golaszewski <bartosz.golaszewski@...aro.org>,
	Bartosz Golaszewski <brgl@...ev.pl>, linux-pci@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Dmitry Baryshkov <dmitry.baryshkov@...aro.org>,
	Johan Hovold <johan+linaro@...nel.org>,
	Abel Vesa <abel.vesa@...aro.org>,
	Stephan Gerhold <stephan.gerhold@...aro.org>,
	Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
	Bjorn Andersson <bjorn.andersson@....qualcomm.com>,
	stable+noautosel@...nel.org,
	Krishna chaitanya chundru <quic_krichai@...cinc.com>
Subject: Re: [PATCH v2 3/5] PCI/pwrctl: Ensure that the pwrctl drivers are
 probed before the PCI client drivers

On Wed, Nov 20, 2024 at 10:32:32PM +0530, Manivannan Sadhasivam wrote:
> On Wed, Nov 20, 2024 at 10:10:47AM -0600, Bjorn Helgaas wrote:
> > On Fri, Oct 25, 2024 at 01:24:53PM +0530, Manivannan Sadhasivam via B4 Relay wrote:
> > > From: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
> > > 
> > > As per the kernel device driver model, pwrctl device is the supplier for
> > > the PCI device. But the device link that enforces the supplier-consumer
> > > relationship is created inside the pwrctl driver currently. Due to this,
> > > the driver model doesn't prevent probing of the PCI client drivers before
> > > probing the corresponding pwrctl drivers. This may lead to a race condition
> > > if the PCI device was already powered on by the bootloader (before the
> > > pwrctl driver).
> > 
> > > +	 * Create a device link between the PCI device and pwrctl device (if
> > > +	 * exists). This ensures that the pwrctl drivers are probed before the
> > > +	 * PCI client drivers.
> > > +	 */
> > > +	pdev = of_find_device_by_node(dn);
> > > +	if (pdev) {
> > > +		if (!device_link_add(&dev->dev, &pdev->dev, DL_FLAG_AUTOREMOVE_CONSUMER))
> > > +			pci_err(dev, "failed to add device link between %s and %s\n",
> > > +				dev_name(&dev->dev), pdev->name);
> > 
> > This prints the name for "dev" twice (once by pci_err(dev) and again
> > from dev_name(&dev->dev)).  Is it helpful to see it twice here?
> 
> Hmm, not very much. It could be reworded as below:
> 
> 	pci_err(dev, "failed to link: %s\n", pdev->name);

OK.  I updated the comment and the message like this (also renamed
of_pci_is_supply_present() to of_pci_supply_present() so it reads more
naturally in "if" statements):

-	 * Create a device link between the PCI device and pwrctrl device (if
-	 * exists). This ensures that the pwrctrl drivers are probed before the
-	 * PCI client drivers.
+	 * If the PCI device is associated with a pwrctrl device with a
+	 * power supply, create a device link between the PCI device and
+	 * pwrctrl device.  This ensures that pwrctrl drivers are probed
+	 * before PCI client drivers.
 	 */
 	pdev = of_find_device_by_node(dn);
-	if (pdev) {
+	if (pdev && of_pci_supply_present(dn)) {
 		if (!device_link_add(&dev->dev, &pdev->dev,
 				     DL_FLAG_AUTOREMOVE_CONSUMER))
-			pci_err(dev, "failed to add device link between %s and %s\n",
-				dev_name(&dev->dev), pdev->name);
+			pci_err(dev, "failed to add device link to power control device %s\n",
+				pdev->name);
 	}