| Message ID | 515c9685-bad2-86e0-1be6-f9f63578b864@gmail.com |
|---|---|
| State | Accepted |
| Commit | c073b25dad0c2668b73ae9e8652bc3b8012e762b |
| Headers | show |
| Series | [RESEND] i2c: i801: Stop using pm_runtime_set_autosuspend_delay(-1) | expand |
On Tue, Sep 7, 2021 at 10:33 PM Heiner Kallweit <hkallweit1@gmail.com> wrote: > > The original change works as intended, but Andy pointed in [0] that now > userspace could re-enable RPM via sysfs, even though we explicitly want > to disable it. So effectively revert the original patch, just with small > improvements: > - Calls to pm_runtime_allow()/pm_runtime_forbid() don't have to be > balanced, so we can remove the call to pm_runtime_forbid() in > i801_remove(). > - priv->acpi_reserved is accessed after i801_acpi_remove(), and according > to Robert [1] the custom handler can't run any longer. Therefore we > don't have to take priv->acpi_lock. > > [0] https://www.spinics.net/lists/linux-i2c/msg52730.html > [1] https://lore.kernel.org/linux-acpi/BYAPR11MB32561D19A0FD9AB93E2B1E5287D39@BYAPR11MB3256.namprd11.prod.outlook.com/T/#t > > Fixes: 4e60d5dd10cd ("i2c: i801: Improve disabling runtime pm") > Reported-by: Andy Shevchenko <andriy.shevchenko@intel.com> > Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Acked-by: Rafael J. Wysocki <rafael@kernel.org> > --- > - Resend because I missed to cc linux-i2c list > --- > drivers/i2c/busses/i2c-i801.c | 9 +++++---- > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c > index 1f929e6c3..f3c79942c 100644 > --- a/drivers/i2c/busses/i2c-i801.c > +++ b/drivers/i2c/busses/i2c-i801.c > @@ -1623,7 +1623,7 @@ i801_acpi_io_handler(u32 function, acpi_physical_address address, u32 bits, > * BIOS is accessing the host controller so prevent it from > * suspending automatically from now on. > */ > - pm_runtime_set_autosuspend_delay(&pdev->dev, -1); > + pm_runtime_get_sync(&pdev->dev); > } > > if ((function & ACPI_IO_MASK) == ACPI_READ) > @@ -1890,9 +1890,6 @@ static void i801_remove(struct pci_dev *dev) > { > struct i801_priv *priv = pci_get_drvdata(dev); > > - pm_runtime_forbid(&dev->dev); > - pm_runtime_get_noresume(&dev->dev); > - > i801_disable_host_notify(priv); > i801_del_mux(priv); > i2c_del_adapter(&priv->adapter); > @@ -1901,6 +1898,10 @@ static void i801_remove(struct pci_dev *dev) > > platform_device_unregister(priv->tco_pdev); > > + /* if acpi_reserved is set then usage_count is incremented already */ > + if (!priv->acpi_reserved) > + pm_runtime_get_noresume(&dev->dev); > + > /* > * do not call pci_disable_device(dev) since it can cause hard hangs on > * some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010) > -- > 2.33.0 >
Hi Heiner, On Tue, 07 Sep 2021 22:33:02 +0200, Heiner Kallweit wrote: > The original change works as intended, but Andy pointed in [0] that now > userspace could re-enable RPM via sysfs, even though we explicitly want > to disable it. So effectively revert the original patch, just with small > improvements: > - Calls to pm_runtime_allow()/pm_runtime_forbid() don't have to be > balanced, so we can remove the call to pm_runtime_forbid() in > i801_remove(). > - priv->acpi_reserved is accessed after i801_acpi_remove(), and according > to Robert [1] the custom handler can't run any longer. Therefore we > don't have to take priv->acpi_lock. > > [0] https://www.spinics.net/lists/linux-i2c/msg52730.html > [1] https://lore.kernel.org/linux-acpi/BYAPR11MB32561D19A0FD9AB93E2B1E5287D39@BYAPR11MB3256.namprd11.prod.outlook.com/T/#t > > Fixes: 4e60d5dd10cd ("i2c: i801: Improve disabling runtime pm") > Reported-by: Andy Shevchenko <andriy.shevchenko@intel.com> > Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> > --- > - Resend because I missed to cc linux-i2c list > --- > drivers/i2c/busses/i2c-i801.c | 9 +++++---- > 1 file changed, 5 insertions(+), 4 deletions(-) > > diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c > index 1f929e6c3..f3c79942c 100644 > --- a/drivers/i2c/busses/i2c-i801.c > +++ b/drivers/i2c/busses/i2c-i801.c > @@ -1623,7 +1623,7 @@ i801_acpi_io_handler(u32 function, acpi_physical_address address, u32 bits, > * BIOS is accessing the host controller so prevent it from > * suspending automatically from now on. > */ > - pm_runtime_set_autosuspend_delay(&pdev->dev, -1); > + pm_runtime_get_sync(&pdev->dev); > } > > if ((function & ACPI_IO_MASK) == ACPI_READ) > @@ -1890,9 +1890,6 @@ static void i801_remove(struct pci_dev *dev) > { > struct i801_priv *priv = pci_get_drvdata(dev); > > - pm_runtime_forbid(&dev->dev); > - pm_runtime_get_noresume(&dev->dev); > - > i801_disable_host_notify(priv); > i801_del_mux(priv); > i2c_del_adapter(&priv->adapter); > @@ -1901,6 +1898,10 @@ static void i801_remove(struct pci_dev *dev) > > platform_device_unregister(priv->tco_pdev); > > + /* if acpi_reserved is set then usage_count is incremented already */ > + if (!priv->acpi_reserved) > + pm_runtime_get_noresume(&dev->dev); > + > /* > * do not call pci_disable_device(dev) since it can cause hard hangs on > * some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010) Thanks for following up. I'm not familiar with power management so I'll trust you, Andy and Robert on that. Reviewed-by: Jean Delvare <jdelvare@suse.de> -- Jean Delvare SUSE L3 Support
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c index 1f929e6c3..f3c79942c 100644 --- a/drivers/i2c/busses/i2c-i801.c +++ b/drivers/i2c/busses/i2c-i801.c @@ -1623,7 +1623,7 @@ i801_acpi_io_handler(u32 function, acpi_physical_address address, u32 bits, * BIOS is accessing the host controller so prevent it from * suspending automatically from now on. */ - pm_runtime_set_autosuspend_delay(&pdev->dev, -1); + pm_runtime_get_sync(&pdev->dev); } if ((function & ACPI_IO_MASK) == ACPI_READ) @@ -1890,9 +1890,6 @@ static void i801_remove(struct pci_dev *dev) { struct i801_priv *priv = pci_get_drvdata(dev); - pm_runtime_forbid(&dev->dev); - pm_runtime_get_noresume(&dev->dev); - i801_disable_host_notify(priv); i801_del_mux(priv); i2c_del_adapter(&priv->adapter); @@ -1901,6 +1898,10 @@ static void i801_remove(struct pci_dev *dev) platform_device_unregister(priv->tco_pdev); + /* if acpi_reserved is set then usage_count is incremented already */ + if (!priv->acpi_reserved) + pm_runtime_get_noresume(&dev->dev); + /* * do not call pci_disable_device(dev) since it can cause hard hangs on * some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010)
The original change works as intended, but Andy pointed in [0] that now userspace could re-enable RPM via sysfs, even though we explicitly want to disable it. So effectively revert the original patch, just with small improvements: - Calls to pm_runtime_allow()/pm_runtime_forbid() don't have to be balanced, so we can remove the call to pm_runtime_forbid() in i801_remove(). - priv->acpi_reserved is accessed after i801_acpi_remove(), and according to Robert [1] the custom handler can't run any longer. Therefore we don't have to take priv->acpi_lock. [0] https://www.spinics.net/lists/linux-i2c/msg52730.html [1] https://lore.kernel.org/linux-acpi/BYAPR11MB32561D19A0FD9AB93E2B1E5287D39@BYAPR11MB3256.namprd11.prod.outlook.com/T/#t Fixes: 4e60d5dd10cd ("i2c: i801: Improve disabling runtime pm") Reported-by: Andy Shevchenko <andriy.shevchenko@intel.com> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> --- - Resend because I missed to cc linux-i2c list --- drivers/i2c/busses/i2c-i801.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)