linux-kernel - [PATCH] platform/x86: intel_pmc_core: do not create a static struct device

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-next>] [day] [month] [year] [list]

Date:   Wed, 23 Sep 2020 20:48:03 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     rajneesh.bhardwaj@...el.com, vishwanath.somayaji@...el.com
Cc:     platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Darren Hart <dvhart@...radead.org>,
        Andy Shevchenko <andy@...radead.org>,
        Rajat Jain <rajatja@...gle.com>,
        Maximilian Luz <luzmaximilian@...il.com>
Subject: [PATCH] platform/x86: intel_pmc_core: do not create a static struct device

A struct device is a dynamic structure, with reference counting.
"Tricking" the kernel to make a dynamic structure static, by working
around the driver core release detection logic, is not nice.

Because of this, this code has been used as an example for others on
"how to do things", which is just about the worst thing possible to have
happen.

Fix this all up by making the platform device dynamic and providing a
real release function.

Cc: Rajneesh Bhardwaj <rajneesh.bhardwaj@...el.com>
Cc: Vishwanath Somayaji <vishwanath.somayaji@...el.com>
Cc: Darren Hart <dvhart@...radead.org>
Cc: Andy Shevchenko <andy@...radead.org>
Cc: Rajat Jain <rajatja@...gle.com>
Cc: platform-driver-x86@...r.kernel.org
Cc: linux-kernel@...r.kernel.org
Reported-by: Maximilian Luz <luzmaximilian@...il.com>
Fixes: b02f6a2ef0a1 ("platform/x86: intel_pmc_core: Attach using APCI HID "INT33A1"")
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
 drivers/platform/x86/intel_pmc_core_pltdrv.c | 26 +++++++++++++-------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/drivers/platform/x86/intel_pmc_core_pltdrv.c b/drivers/platform/x86/intel_pmc_core_pltdrv.c
index 731281855cc8..73797680b895 100644
--- a/drivers/platform/x86/intel_pmc_core_pltdrv.c
+++ b/drivers/platform/x86/intel_pmc_core_pltdrv.c
@@ -20,15 +20,10 @@
 
 static void intel_pmc_core_release(struct device *dev)
 {
-	/* Nothing to do. */
+	kfree(dev);
 }
 
-static struct platform_device pmc_core_device = {
-	.name = "intel_pmc_core",
-	.dev  = {
-		.release = intel_pmc_core_release,
-	},
-};
+static struct platform_device *pmc_core_device;
 
 /*
  * intel_pmc_core_platform_ids is the list of platforms where we want to
@@ -52,6 +47,8 @@ MODULE_DEVICE_TABLE(x86cpu, intel_pmc_core_platform_ids);
 
 static int __init pmc_core_platform_init(void)
 {
+	int retval;
+
 	/* Skip creating the platform device if ACPI already has a device */
 	if (acpi_dev_present("INT33A1", NULL, -1))
 		return -ENODEV;
@@ -59,12 +56,23 @@ static int __init pmc_core_platform_init(void)
 	if (!x86_match_cpu(intel_pmc_core_platform_ids))
 		return -ENODEV;
 
-	return platform_device_register(&pmc_core_device);
+	pmc_core_device = kzalloc(sizeof(*pmc_core_device), GFP_KERNEL);
+	if (!pmc_core_device)
+		return -ENOMEM;
+
+	pmc_core_device->name = "intel_pmc_core";
+	pmc_core_device->dev.release = intel_pmc_core_release;
+
+	retval = platform_device_register(pmc_core_device);
+	if (retval)
+		kfree(pmc_core_device);
+
+	return retval;
 }
 
 static void __exit pmc_core_platform_exit(void)
 {
-	platform_device_unregister(&pmc_core_device);
+	platform_device_unregister(pmc_core_device);
 }
 
 module_init(pmc_core_platform_init);
-- 
2.28.0