| Message ID | 20200923184803.192265-1-gregkh@linuxfoundation.org (mailing list archive) |
|---|---|
| State | Accepted, archived |
| Delegated to: | Andy Shevchenko |
| Headers | show |
| Series | platform/x86: intel_pmc_core: do not create a static struct device | expand |
On Wed, Sep 23, 2020 at 11:47 AM Greg Kroah-Hartman <gregkh@linuxfoundation.org> wrote: > > 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@intel.com> > Cc: Vishwanath Somayaji <vishwanath.somayaji@intel.com> > Cc: Darren Hart <dvhart@infradead.org> > Cc: Andy Shevchenko <andy@infradead.org> > Cc: Rajat Jain <rajatja@google.com> > Cc: platform-driver-x86@vger.kernel.org > Cc: linux-kernel@vger.kernel.org > Reported-by: Maximilian Luz <luzmaximilian@gmail.com> > Fixes: b02f6a2ef0a1 ("platform/x86: intel_pmc_core: Attach using APCI HID "INT33A1"") > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Acked-by: Rajat Jain <rajatja@google.com> > --- > 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 >
Hi, On 9/23/20 8:48 PM, Greg Kroah-Hartman wrote: > 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@intel.com> > Cc: Vishwanath Somayaji <vishwanath.somayaji@intel.com> > Cc: Darren Hart <dvhart@infradead.org> > Cc: Andy Shevchenko <andy@infradead.org> > Cc: Rajat Jain <rajatja@google.com> > Cc: platform-driver-x86@vger.kernel.org > Cc: linux-kernel@vger.kernel.org > Reported-by: Maximilian Luz <luzmaximilian@gmail.com> > Fixes: b02f6a2ef0a1 ("platform/x86: intel_pmc_core: Attach using APCI HID "INT33A1"") > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Patch looks good to me: Reviewed-by: Hans de Goede <hdegoede@redhat.com> Regards, Hans > --- > 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); >
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);
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@intel.com> Cc: Vishwanath Somayaji <vishwanath.somayaji@intel.com> Cc: Darren Hart <dvhart@infradead.org> Cc: Andy Shevchenko <andy@infradead.org> Cc: Rajat Jain <rajatja@google.com> Cc: platform-driver-x86@vger.kernel.org Cc: linux-kernel@vger.kernel.org Reported-by: Maximilian Luz <luzmaximilian@gmail.com> Fixes: b02f6a2ef0a1 ("platform/x86: intel_pmc_core: Attach using APCI HID "INT33A1"") Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> --- drivers/platform/x86/intel_pmc_core_pltdrv.c | 26 +++++++++++++------- 1 file changed, 17 insertions(+), 9 deletions(-)