[PATCH v1 0/4] drivers/staging: use generic power management
Linux Kernel Mentee: Remove Legacy Power Management. The purpose of this patch series is to remove legacy power management callbacks from amd ethernet drivers. The callbacks performing suspend() and resume() operations are still calling pci_save_state(), pci_set_power_state(), etc. and handling the power management themselves, which is not recommended. The conversion requires the removal of the those function calls and change the callback definition accordingly and make use of dev_pm_ops structure. All patches are compile-tested only. Vaibhav Gupta (4): qlge/qlge_main.c: use genric power management staging/rtl8192e: use generic power management rts5208/rtsx.c: use generic power management vt6655/device_main.c: use generic power management drivers/staging/qlge/qlge_main.c | 36 +--- drivers/staging/rtl8192e/rtl8192e/rtl_core.c | 5 +-- drivers/staging/rtl8192e/rtl8192e/rtl_pm.c | 26 -- drivers/staging/rtl8192e/rtl8192e/rtl_pm.h | 4 +-- drivers/staging/rts5208/rtsx.c | 30 +--- drivers/staging/vt6655/device_main.c | 25 -- 6 files changed, 36 insertions(+), 90 deletions(-) -- 2.27.0 ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v1 2/4] staging/rtl8192e: use generic power management
The structure of working of PM hooks for source files is:
drivers/staging/rtl8192e/rtl8192e/rtl_pm.h : callbacks declared
drivers/staging/rtl8192e/rtl8192e/rtl_pm.c : callbacks defined
drivers/staging/rtl8192e/rtl8192e/rtl_core.c : callbacks used
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead. Use device_set_wakeup_enable() where WOL
is decided by the value of a variable during runtime.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/rtl8192e/rtl8192e/rtl_core.c | 5 ++--
drivers/staging/rtl8192e/rtl8192e/rtl_pm.c | 26 ++--
drivers/staging/rtl8192e/rtl8192e/rtl_pm.h | 4 +--
3 files changed, 12 insertions(+), 23 deletions(-)
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
index a7cd4de65b28..dbcb8d0d9707 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
@@ -63,13 +63,14 @@ static int _rtl92e_pci_probe(struct pci_dev *pdev,
static void _rtl92e_pci_disconnect(struct pci_dev *pdev);
static irqreturn_t _rtl92e_irq(int irq, void *netdev);
+static SIMPLE_DEV_PM_OPS(rtl92e_pm_ops, rtl92e_suspend, rtl92e_resume);
+
static struct pci_driver rtl8192_pci_driver = {
.name = DRV_NAME, /* Driver name */
.id_table = rtl8192_pci_id_tbl, /* PCI_ID table */
.probe = _rtl92e_pci_probe,/* probe fn */
.remove = _rtl92e_pci_disconnect, /* remove fn */
- .suspend = rtl92e_suspend, /* PM suspend fn */
- .resume = rtl92e_resume, /* PM resume fn */
+ .driver.pm = &rtl92e_pm_ops,
};
static short _rtl92e_is_tx_queue_empty(struct net_device *dev);
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
index cd3e17b41d6f..5575186caebd 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
@@ -10,9 +10,9 @@
#include "rtl_pm.h"
-int rtl92e_suspend(struct pci_dev *pdev, pm_message_t state)
+int rtl92e_suspend(struct device *dev_d)
{
- struct net_device *dev = pci_get_drvdata(pdev);
+ struct net_device *dev = dev_get_drvdata(dev_d);
struct r8192_priv *priv = rtllib_priv(dev);
u32 ulRegRead;
@@ -46,40 +46,28 @@ int rtl92e_suspend(struct pci_dev *pdev, pm_message_t state)
out_pci_suspend:
netdev_info(dev, "WOL is %s\n", priv->rtllib->bSupportRemoteWakeUp ?
"Supported" : "Not supported");
- pci_save_state(pdev);
- pci_disable_device(pdev);
- pci_enable_wake(pdev, pci_choose_state(pdev, state),
- priv->rtllib->bSupportRemoteWakeUp ? 1 : 0);
- pci_set_power_state(pdev, pci_choose_state(pdev, state));
+ device_set_wakeup_enable(dev_d, priv->rtllib->bSupportRemoteWakeUp);
mdelay(20);
return 0;
}
-int rtl92e_resume(struct pci_dev *pdev)
+int rtl92e_resume(struct device *dev_d)
{
- struct net_device *dev = pci_get_drvdata(pdev);
+ struct pci_dev *pdev = to_pci_dev(dev_d);
+ struct net_device *dev = dev_get_drvdata(dev_d);
struct r8192_priv *priv = rtllib_priv(dev);
- int err;
u32 val;
netdev_info(dev, ">r8192E resume call.\n");
- pci_set_power_state(pdev, PCI_D0);
-
- err = pci_enable_device(pdev);
- if (err) {
- netdev_err(dev, "pci_enable_device failed on resume\n");
- return err;
- }
- pci_restore_state(pdev);
pci_read_config_dword(pdev, 0x40, &val);
if ((val & 0xff00) != 0)
pci_write_config_dword(pdev, 0x40, val & 0x00ff);
- pci_enable_wake(pdev, PCI_D0, 0);
+ device_wakeup_disable(dev_d);
if (priv->polling_timer_on == 0)
rtl92e_check_rfctrl_gpio_timer(&priv->gpio_polling_timer);
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
index e58f2bcdb1dd..fd8611495975 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
@@ -10,7 +10,7 @@
#include
#include
-int rtl92e_suspend(struct pci_dev *dev, pm_message_t state);
-int rtl92e_resume(struct pci_dev *dev);
+int rtl92e_suspend(struct device *dev_d);
+int rtl92e_resume(struct device *dev_d);
#endif
--
2
[PATCH v1 4/4] vt6655/device_main.c: use generic power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/vt6655/device_main.c | 25 +++--
1 file changed, 7 insertions(+), 18 deletions(-)
diff --git a/drivers/staging/vt6655/device_main.c
b/drivers/staging/vt6655/device_main.c
index 41cbec4134b0..76de1fd568eb 100644
--- a/drivers/staging/vt6655/device_main.c
+++ b/drivers/staging/vt6655/device_main.c
@@ -1766,48 +1766,37 @@ vt6655_probe(struct pci_dev *pcid, const struct
pci_device_id *ent)
/*--*/
-#ifdef CONFIG_PM
-static int vt6655_suspend(struct pci_dev *pcid, pm_message_t state)
+static int __maybe_unused vt6655_suspend(struct device *dev_d)
{
- struct vnt_private *priv = pci_get_drvdata(pcid);
+ struct vnt_private *priv = dev_get_drvdata(dev_d);
unsigned long flags;
spin_lock_irqsave(&priv->lock, flags);
- pci_save_state(pcid);
-
MACbShutdown(priv);
- pci_disable_device(pcid);
-
spin_unlock_irqrestore(&priv->lock, flags);
- pci_set_power_state(pcid, pci_choose_state(pcid, state));
-
return 0;
}
-static int vt6655_resume(struct pci_dev *pcid)
+static int __maybe_unused vt6655_resume(struct device *dev_d)
{
- pci_set_power_state(pcid, PCI_D0);
- pci_enable_wake(pcid, PCI_D0, 0);
- pci_restore_state(pcid);
+ device_wakeup_disable(dev_d);
return 0;
}
-#endif
MODULE_DEVICE_TABLE(pci, vt6655_pci_id_table);
+static SIMPLE_DEV_PM_OPS(vt6655_pm_ops, vt6655_suspend, vt6655_resume);
+
static struct pci_driver device_driver = {
.name = DEVICE_NAME,
.id_table = vt6655_pci_id_table,
.probe = vt6655_probe,
.remove = vt6655_remove,
-#ifdef CONFIG_PM
- .suspend = vt6655_suspend,
- .resume = vt6655_resume,
-#endif
+ .driver.pm = &vt6655_pm_ops,
};
module_pci_driver(device_driver);
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v1 1/4] qlge/qlge_main.c: use genric power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/qlge/qlge_main.c | 36
1 file changed, 9 insertions(+), 27 deletions(-)
diff --git a/drivers/staging/qlge/qlge_main.c b/drivers/staging/qlge/qlge_main.c
index 402edaeffe12..b6f6f681c77b 100644
--- a/drivers/staging/qlge/qlge_main.c
+++ b/drivers/staging/qlge/qlge_main.c
@@ -4763,9 +4763,9 @@ static const struct pci_error_handlers qlge_err_handler =
{
.resume = qlge_io_resume,
};
-static int qlge_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused qlge_suspend(struct device *dev_d)
{
- struct net_device *ndev = pci_get_drvdata(pdev);
+ struct net_device *ndev = dev_get_drvdata(dev_d);
struct ql_adapter *qdev = netdev_priv(ndev);
int err;
@@ -4779,35 +4779,19 @@ static int qlge_suspend(struct pci_dev *pdev,
pm_message_t state)
}
ql_wol(qdev);
- err = pci_save_state(pdev);
- if (err)
- return err;
-
- pci_disable_device(pdev);
-
- pci_set_power_state(pdev, pci_choose_state(pdev, state));
return 0;
}
-#ifdef CONFIG_PM
-static int qlge_resume(struct pci_dev *pdev)
+static int __maybe_unused qlge_resume(struct device *dev_d)
{
- struct net_device *ndev = pci_get_drvdata(pdev);
+ struct net_device *ndev = dev_get_drvdata(dev_d);
struct ql_adapter *qdev = netdev_priv(ndev);
int err;
- pci_set_power_state(pdev, PCI_D0);
- pci_restore_state(pdev);
- err = pci_enable_device(pdev);
- if (err) {
- netif_err(qdev, ifup, qdev->ndev, "Cannot enable PCI device
from suspend\n");
- return err;
- }
pci_set_master(pdev);
- pci_enable_wake(pdev, PCI_D3hot, 0);
- pci_enable_wake(pdev, PCI_D3cold, 0);
+ device_wakeup_disable(dev_d);
if (netif_running(ndev)) {
err = ql_adapter_up(qdev);
@@ -4820,22 +4804,20 @@ static int qlge_resume(struct pci_dev *pdev)
return 0;
}
-#endif /* CONFIG_PM */
static void qlge_shutdown(struct pci_dev *pdev)
{
- qlge_suspend(pdev, PMSG_SUSPEND);
+ qlge_suspend(&pdev->dev);
}
+static SIMPLE_DEV_PM_OPS(qlge_pm_ops, qlge_suspend, qlge_resume);
+
static struct pci_driver qlge_driver = {
.name = DRV_NAME,
.id_table = qlge_pci_tbl,
.probe = qlge_probe,
.remove = qlge_remove,
-#ifdef CONFIG_PM
- .suspend = qlge_suspend,
- .resume = qlge_resume,
-#endif
+ .driver.pm = &qlge_pm_ops,
.shutdown = qlge_shutdown,
.err_handler = &qlge_err_handler
};
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v1 3/4] rts5208/rtsx.c: use generic power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/rts5208/rtsx.c | 30 --
1 file changed, 8 insertions(+), 22 deletions(-)
diff --git a/drivers/staging/rts5208/rtsx.c b/drivers/staging/rts5208/rtsx.c
index be0053c795b7..6ca90694db8b 100644
--- a/drivers/staging/rts5208/rtsx.c
+++ b/drivers/staging/rts5208/rtsx.c
@@ -258,12 +258,12 @@ static int rtsx_acquire_irq(struct rtsx_dev *dev)
return 0;
}
-#ifdef CONFIG_PM
/*
* power management
*/
-static int rtsx_suspend(struct pci_dev *pci, pm_message_t state)
+static int __maybe_unused rtsx_suspend(struct device *dev_d)
{
+ struct pci_dev *pci = to_pci_dev(dev_d);
struct rtsx_dev *dev = pci_get_drvdata(pci);
struct rtsx_chip *chip;
@@ -285,10 +285,7 @@ static int rtsx_suspend(struct pci_dev *pci, pm_message_t
state)
if (chip->msi_en)
pci_disable_msi(pci);
- pci_save_state(pci);
- pci_enable_wake(pci, pci_choose_state(pci, state), 1);
- pci_disable_device(pci);
- pci_set_power_state(pci, pci_choose_state(pci, state));
+ device_wakeup_enable(dev_d);
/* unlock the device pointers */
mutex_unlock(&dev->dev_mutex);
@@ -296,8 +293,9 @@ static int rtsx_suspend(struct pci_dev *pci, pm_message_t
state)
return 0;
}
-static int rtsx_resume(struct pci_dev *pci)
+static int __maybe_unused rtsx_resume(struct device *dev_d)
{
+ struct pci_dev *pci = to_pci_dev(dev_d);
struct rtsx_dev *dev = pci_get_drvdata(pci);
struct rtsx_chip *chip;
@@ -309,16 +307,6 @@ static int rtsx_resume(struct pci_dev *pci)
/* lock the device pointers */
mutex_lock(&dev->dev_mutex);
- pci_set_power_state(pci, PCI_D0);
- pci_restore_state(pci);
- if (pci_enable_device(pci) < 0) {
- dev_err(&dev->pci->dev,
- "%s: pci_enable_device failed, disabling device\n",
- CR_DRIVER_NAME);
- /* unlock the device pointers */
- mutex_unlock(&dev->dev_mutex);
- return -EIO;
- }
pci_set_master(pci);
if (chip->msi_en) {
@@ -340,7 +328,6 @@ static int rtsx_resume(struct pci_dev *pci)
return 0;
}
-#endif /* CONFIG_PM */
static void rtsx_shutdown(struct pci_dev *pci)
{
@@ -999,16 +986,15 @@ static const struct pci_device_id rtsx_ids[] = {
MODULE_DEVICE_TABLE(pci, rtsx_ids);
+static SIMPLE_DEV_PM_OPS(rtsx_pm_ops, rtsx_suspend, rtsx_resume);
+
/* pci_driver definition */
static struct pci_driver rtsx_driver = {
.name = CR_DRIVER_NAME,
.id_table = rtsx_ids,
.probe = rtsx_probe,
.remove = rtsx_remove,
-#ifdef CONFIG_PM
- .suspend = rtsx_suspend,
- .resume = rtsx_resume,
-#endif
+ .driver.pm = &rtsx_pm_ops,
.shutdown = rtsx_shutdown,
};
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
Re: [PATCH v1 1/4] qlge/qlge_main.c: use genric power management
On Mon, 29 Jun 2020 at 21:39, kernel test robot wrote:
>
> Hi Vaibhav,
>
> Thank you for the patch! Yet something to improve:
>
> [auto build test ERROR on staging/staging-testing]
> [also build test ERROR on v5.8-rc3 next-20200629]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use as documented in
> https://git-scm.com/docs/git-format-patch]
>
> url:
> https://github.com/0day-ci/linux/commits/Vaibhav-Gupta/drivers-staging-use-generic-power-management/20200629-163141
> base: https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git
> 347fa58ff5558075eec98725029c443c80ffbf4a
> config: x86_64-rhel-7.6 (attached as .config)
> compiler: gcc-9 (Debian 9.3.0-13) 9.3.0
> reproduce (this is a W=1 build):
> # save the attached .config to linux build tree
> make W=1 ARCH=x86_64
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot
>
> All errors (new ones prefixed by >>):
>
>drivers/staging/qlge/qlge_main.c: In function 'qlge_resume':
> >> drivers/staging/qlge/qlge_main.c:4793:17: error: 'pdev' undeclared (first
> >> use in this function); did you mean 'qdev'?
> 4793 | pci_set_master(pdev);
> | ^~~~
> | qdev
>drivers/staging/qlge/qlge_main.c:4793:17: note: each undeclared identifier
> is reported only once for each function it appears in
>
> vim +4793 drivers/staging/qlge/qlge_main.c
>
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4786
> 1aed508211e137 drivers/staging/qlge/qlge_main.c Vaibhav Gupta 2020-06-29
> 4787 static int __maybe_unused qlge_resume(struct device *dev_d)
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4788 {
> 1aed508211e137 drivers/staging/qlge/qlge_main.c Vaibhav Gupta 2020-06-29
> 4789 struct net_device *ndev = dev_get_drvdata(dev_d);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4790 struct ql_adapter *qdev = netdev_priv(ndev);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4791 int err;
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4792
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer 2008-09-18
> @4793 pci_set_master(pdev);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4794
> 1aed508211e137 drivers/staging/qlge/qlge_main.c Vaibhav Gupta 2020-06-29
> 4795 device_wakeup_disable(dev_d);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4796
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4797 if (netif_running(ndev)) {
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4798 err = ql_adapter_up(qdev);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4799 if (err)
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4800 return err;
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4801 }
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4802
> 72046d84f0d6e3 drivers/net/qlge/qlge_main.c Breno Leitao 2010-07-01
> 4803 mod_timer(&qdev->timer, jiffies + (5 * HZ));
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4804 netif_device_attach(ndev);
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4805
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4806 return 0;
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18
> 4807 }
> c4e84bde1d595d drivers/net/qlge/qlge_main.c Ron Mercer2008-09-18 4808
>
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-...@lists.01.org
This is a genuine error. I remember taking care of it when I upgraded
the driver.
Must have been caused by some unnoticed reverts at my side.
I will fix it and send a v2 patch-set. Thanks for reporting.
-- Vaibhav gupta
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v2 1/4] qlge/qlge_main.c: use generic power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/qlge/qlge_main.c | 38 +---
1 file changed, 10 insertions(+), 28 deletions(-)
diff --git a/drivers/staging/qlge/qlge_main.c b/drivers/staging/qlge/qlge_main.c
index 402edaeffe12..86d1b53da2c2 100644
--- a/drivers/staging/qlge/qlge_main.c
+++ b/drivers/staging/qlge/qlge_main.c
@@ -4763,9 +4763,9 @@ static const struct pci_error_handlers qlge_err_handler =
{
.resume = qlge_io_resume,
};
-static int qlge_suspend(struct pci_dev *pdev, pm_message_t state)
+static int __maybe_unused qlge_suspend(struct device *dev_d)
{
- struct net_device *ndev = pci_get_drvdata(pdev);
+ struct net_device *ndev = dev_get_drvdata(dev_d);
struct ql_adapter *qdev = netdev_priv(ndev);
int err;
@@ -4779,35 +4779,19 @@ static int qlge_suspend(struct pci_dev *pdev,
pm_message_t state)
}
ql_wol(qdev);
- err = pci_save_state(pdev);
- if (err)
- return err;
-
- pci_disable_device(pdev);
-
- pci_set_power_state(pdev, pci_choose_state(pdev, state));
return 0;
}
-#ifdef CONFIG_PM
-static int qlge_resume(struct pci_dev *pdev)
+static int __maybe_unused qlge_resume(struct device *dev_d)
{
- struct net_device *ndev = pci_get_drvdata(pdev);
+ struct net_device *ndev = dev_get_drvdata(dev_d);
struct ql_adapter *qdev = netdev_priv(ndev);
int err;
- pci_set_power_state(pdev, PCI_D0);
- pci_restore_state(pdev);
- err = pci_enable_device(pdev);
- if (err) {
- netif_err(qdev, ifup, qdev->ndev, "Cannot enable PCI device
from suspend\n");
- return err;
- }
- pci_set_master(pdev);
+ pci_set_master(to_pci_dev(dev_d));
- pci_enable_wake(pdev, PCI_D3hot, 0);
- pci_enable_wake(pdev, PCI_D3cold, 0);
+ device_wakeup_disable(dev_d);
if (netif_running(ndev)) {
err = ql_adapter_up(qdev);
@@ -4820,22 +4804,20 @@ static int qlge_resume(struct pci_dev *pdev)
return 0;
}
-#endif /* CONFIG_PM */
static void qlge_shutdown(struct pci_dev *pdev)
{
- qlge_suspend(pdev, PMSG_SUSPEND);
+ qlge_suspend(&pdev->dev);
}
+static SIMPLE_DEV_PM_OPS(qlge_pm_ops, qlge_suspend, qlge_resume);
+
static struct pci_driver qlge_driver = {
.name = DRV_NAME,
.id_table = qlge_pci_tbl,
.probe = qlge_probe,
.remove = qlge_remove,
-#ifdef CONFIG_PM
- .suspend = qlge_suspend,
- .resume = qlge_resume,
-#endif
+ .driver.pm = &qlge_pm_ops,
.shutdown = qlge_shutdown,
.err_handler = &qlge_err_handler
};
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v2 4/4] vt6655/device_main.c: use generic power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/vt6655/device_main.c | 25 +++--
1 file changed, 7 insertions(+), 18 deletions(-)
diff --git a/drivers/staging/vt6655/device_main.c
b/drivers/staging/vt6655/device_main.c
index 41cbec4134b0..76de1fd568eb 100644
--- a/drivers/staging/vt6655/device_main.c
+++ b/drivers/staging/vt6655/device_main.c
@@ -1766,48 +1766,37 @@ vt6655_probe(struct pci_dev *pcid, const struct
pci_device_id *ent)
/*--*/
-#ifdef CONFIG_PM
-static int vt6655_suspend(struct pci_dev *pcid, pm_message_t state)
+static int __maybe_unused vt6655_suspend(struct device *dev_d)
{
- struct vnt_private *priv = pci_get_drvdata(pcid);
+ struct vnt_private *priv = dev_get_drvdata(dev_d);
unsigned long flags;
spin_lock_irqsave(&priv->lock, flags);
- pci_save_state(pcid);
-
MACbShutdown(priv);
- pci_disable_device(pcid);
-
spin_unlock_irqrestore(&priv->lock, flags);
- pci_set_power_state(pcid, pci_choose_state(pcid, state));
-
return 0;
}
-static int vt6655_resume(struct pci_dev *pcid)
+static int __maybe_unused vt6655_resume(struct device *dev_d)
{
- pci_set_power_state(pcid, PCI_D0);
- pci_enable_wake(pcid, PCI_D0, 0);
- pci_restore_state(pcid);
+ device_wakeup_disable(dev_d);
return 0;
}
-#endif
MODULE_DEVICE_TABLE(pci, vt6655_pci_id_table);
+static SIMPLE_DEV_PM_OPS(vt6655_pm_ops, vt6655_suspend, vt6655_resume);
+
static struct pci_driver device_driver = {
.name = DEVICE_NAME,
.id_table = vt6655_pci_id_table,
.probe = vt6655_probe,
.remove = vt6655_remove,
-#ifdef CONFIG_PM
- .suspend = vt6655_suspend,
- .resume = vt6655_resume,
-#endif
+ .driver.pm = &vt6655_pm_ops,
};
module_pci_driver(device_driver);
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v2 3/4] rts5208/rtsx.c: use generic power management
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/rts5208/rtsx.c | 30 --
1 file changed, 8 insertions(+), 22 deletions(-)
diff --git a/drivers/staging/rts5208/rtsx.c b/drivers/staging/rts5208/rtsx.c
index be0053c795b7..6ca90694db8b 100644
--- a/drivers/staging/rts5208/rtsx.c
+++ b/drivers/staging/rts5208/rtsx.c
@@ -258,12 +258,12 @@ static int rtsx_acquire_irq(struct rtsx_dev *dev)
return 0;
}
-#ifdef CONFIG_PM
/*
* power management
*/
-static int rtsx_suspend(struct pci_dev *pci, pm_message_t state)
+static int __maybe_unused rtsx_suspend(struct device *dev_d)
{
+ struct pci_dev *pci = to_pci_dev(dev_d);
struct rtsx_dev *dev = pci_get_drvdata(pci);
struct rtsx_chip *chip;
@@ -285,10 +285,7 @@ static int rtsx_suspend(struct pci_dev *pci, pm_message_t
state)
if (chip->msi_en)
pci_disable_msi(pci);
- pci_save_state(pci);
- pci_enable_wake(pci, pci_choose_state(pci, state), 1);
- pci_disable_device(pci);
- pci_set_power_state(pci, pci_choose_state(pci, state));
+ device_wakeup_enable(dev_d);
/* unlock the device pointers */
mutex_unlock(&dev->dev_mutex);
@@ -296,8 +293,9 @@ static int rtsx_suspend(struct pci_dev *pci, pm_message_t
state)
return 0;
}
-static int rtsx_resume(struct pci_dev *pci)
+static int __maybe_unused rtsx_resume(struct device *dev_d)
{
+ struct pci_dev *pci = to_pci_dev(dev_d);
struct rtsx_dev *dev = pci_get_drvdata(pci);
struct rtsx_chip *chip;
@@ -309,16 +307,6 @@ static int rtsx_resume(struct pci_dev *pci)
/* lock the device pointers */
mutex_lock(&dev->dev_mutex);
- pci_set_power_state(pci, PCI_D0);
- pci_restore_state(pci);
- if (pci_enable_device(pci) < 0) {
- dev_err(&dev->pci->dev,
- "%s: pci_enable_device failed, disabling device\n",
- CR_DRIVER_NAME);
- /* unlock the device pointers */
- mutex_unlock(&dev->dev_mutex);
- return -EIO;
- }
pci_set_master(pci);
if (chip->msi_en) {
@@ -340,7 +328,6 @@ static int rtsx_resume(struct pci_dev *pci)
return 0;
}
-#endif /* CONFIG_PM */
static void rtsx_shutdown(struct pci_dev *pci)
{
@@ -999,16 +986,15 @@ static const struct pci_device_id rtsx_ids[] = {
MODULE_DEVICE_TABLE(pci, rtsx_ids);
+static SIMPLE_DEV_PM_OPS(rtsx_pm_ops, rtsx_suspend, rtsx_resume);
+
/* pci_driver definition */
static struct pci_driver rtsx_driver = {
.name = CR_DRIVER_NAME,
.id_table = rtsx_ids,
.probe = rtsx_probe,
.remove = rtsx_remove,
-#ifdef CONFIG_PM
- .suspend = rtsx_suspend,
- .resume = rtsx_resume,
-#endif
+ .driver.pm = &rtsx_pm_ops,
.shutdown = rtsx_shutdown,
};
--
2.27.0
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v2 0/4] drivers/staging: use generic power management
Linux Kernel Mentee: Remove Legacy Power Management. The purpose of this patch series is to remove legacy power management callbacks from amd ethernet drivers. The callbacks performing suspend() and resume() operations are still calling pci_save_state(), pci_set_power_state(), etc. and handling the power management themselves, which is not recommended. The conversion requires the removal of the those function calls and change the callback definition accordingly and make use of dev_pm_ops structure. All patches are compile-tested only. v2: An error was reported by kbuild in v1. pci_set_master() function call in drivers/staging/qlge/qlge_main.c, inside qlge_resume(), was passing argument which got deprecated after upgrade to generic PM. The error is fixed in v2. Vaibhav Gupta (4): qlge/qlge_main.c: use generic power management staging/rtl8192e: use generic power management rts5208/rtsx.c: use generic power management vt6655/device_main.c: use generic power management drivers/staging/qlge/qlge_main.c | 38 ++-- drivers/staging/rtl8192e/rtl8192e/rtl_core.c | 5 +-- drivers/staging/rtl8192e/rtl8192e/rtl_pm.c | 26 -- drivers/staging/rtl8192e/rtl8192e/rtl_pm.h | 4 +-- drivers/staging/rts5208/rtsx.c | 30 +--- drivers/staging/vt6655/device_main.c | 25 - 6 files changed, 37 insertions(+), 91 deletions(-) -- 2.27.0 ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
[PATCH v2 2/4] staging/rtl8192e: use generic power management
The structure of working of PM hooks for source files is:
drivers/staging/rtl8192e/rtl8192e/rtl_pm.h : callbacks declared
drivers/staging/rtl8192e/rtl8192e/rtl_pm.c : callbacks defined
drivers/staging/rtl8192e/rtl8192e/rtl_core.c : callbacks used
Drivers should not use legacy power management as they have to manage power
states and related operations, for the device, themselves. This driver was
handling them with the help of PCI helper functions like
pci_save/restore_state(), pci_enable/disable_device(), etc.
With generic PM, all essentials will be handled by the PCI core. Driver
needs to do only device-specific operations.
The driver was also using pci_enable_wake(...,..., 0) to disable wake. Use
device_wakeup_disable() instead. Use device_set_wakeup_enable() where WOL
is decided by the value of a variable during runtime.
Compile-tested only.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/rtl8192e/rtl8192e/rtl_core.c | 5 ++--
drivers/staging/rtl8192e/rtl8192e/rtl_pm.c | 26 ++--
drivers/staging/rtl8192e/rtl8192e/rtl_pm.h | 4 +--
3 files changed, 12 insertions(+), 23 deletions(-)
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
index a7cd4de65b28..dbcb8d0d9707 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_core.c
@@ -63,13 +63,14 @@ static int _rtl92e_pci_probe(struct pci_dev *pdev,
static void _rtl92e_pci_disconnect(struct pci_dev *pdev);
static irqreturn_t _rtl92e_irq(int irq, void *netdev);
+static SIMPLE_DEV_PM_OPS(rtl92e_pm_ops, rtl92e_suspend, rtl92e_resume);
+
static struct pci_driver rtl8192_pci_driver = {
.name = DRV_NAME, /* Driver name */
.id_table = rtl8192_pci_id_tbl, /* PCI_ID table */
.probe = _rtl92e_pci_probe,/* probe fn */
.remove = _rtl92e_pci_disconnect, /* remove fn */
- .suspend = rtl92e_suspend, /* PM suspend fn */
- .resume = rtl92e_resume, /* PM resume fn */
+ .driver.pm = &rtl92e_pm_ops,
};
static short _rtl92e_is_tx_queue_empty(struct net_device *dev);
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
index cd3e17b41d6f..5575186caebd 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.c
@@ -10,9 +10,9 @@
#include "rtl_pm.h"
-int rtl92e_suspend(struct pci_dev *pdev, pm_message_t state)
+int rtl92e_suspend(struct device *dev_d)
{
- struct net_device *dev = pci_get_drvdata(pdev);
+ struct net_device *dev = dev_get_drvdata(dev_d);
struct r8192_priv *priv = rtllib_priv(dev);
u32 ulRegRead;
@@ -46,40 +46,28 @@ int rtl92e_suspend(struct pci_dev *pdev, pm_message_t state)
out_pci_suspend:
netdev_info(dev, "WOL is %s\n", priv->rtllib->bSupportRemoteWakeUp ?
"Supported" : "Not supported");
- pci_save_state(pdev);
- pci_disable_device(pdev);
- pci_enable_wake(pdev, pci_choose_state(pdev, state),
- priv->rtllib->bSupportRemoteWakeUp ? 1 : 0);
- pci_set_power_state(pdev, pci_choose_state(pdev, state));
+ device_set_wakeup_enable(dev_d, priv->rtllib->bSupportRemoteWakeUp);
mdelay(20);
return 0;
}
-int rtl92e_resume(struct pci_dev *pdev)
+int rtl92e_resume(struct device *dev_d)
{
- struct net_device *dev = pci_get_drvdata(pdev);
+ struct pci_dev *pdev = to_pci_dev(dev_d);
+ struct net_device *dev = dev_get_drvdata(dev_d);
struct r8192_priv *priv = rtllib_priv(dev);
- int err;
u32 val;
netdev_info(dev, ">r8192E resume call.\n");
- pci_set_power_state(pdev, PCI_D0);
-
- err = pci_enable_device(pdev);
- if (err) {
- netdev_err(dev, "pci_enable_device failed on resume\n");
- return err;
- }
- pci_restore_state(pdev);
pci_read_config_dword(pdev, 0x40, &val);
if ((val & 0xff00) != 0)
pci_write_config_dword(pdev, 0x40, val & 0x00ff);
- pci_enable_wake(pdev, PCI_D0, 0);
+ device_wakeup_disable(dev_d);
if (priv->polling_timer_on == 0)
rtl92e_check_rfctrl_gpio_timer(&priv->gpio_polling_timer);
diff --git a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
index e58f2bcdb1dd..fd8611495975 100644
--- a/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
+++ b/drivers/staging/rtl8192e/rtl8192e/rtl_pm.h
@@ -10,7 +10,7 @@
#include
#include
-int rtl92e_suspend(struct pci_dev *dev, pm_message_t state);
-int rtl92e_resume(struct pci_dev *dev);
+int rtl92e_suspend(struct device *dev_d);
+int rtl92e_resume(struct device *dev_d);
#endif
--
2
[PATCH v1] staging: sm750fb: use generic power management
Drivers using legacy power management .suspen()/.resume() callbacks
have to manage PCI states and device's PM states themselves. They also
need to take care of standard configuration registers.
Switch to generic power management framework using a single
"struct dev_pm_ops" variable to take the unnecessary load from the driver.
This also avoids the need for the driver to directly call most of the PCI
helper functions and device power state control functions, as through
the generic framework PCI Core takes care of the necessary operations,
and drivers are required to do only device-specific jobs.
Signed-off-by: Vaibhav Gupta
---
drivers/staging/sm750fb/sm750.c | 91 ++---
1 file changed, 17 insertions(+), 74 deletions(-)
diff --git a/drivers/staging/sm750fb/sm750.c b/drivers/staging/sm750fb/sm750.c
index a1a82e59dfee..84fb585a5739 100644
--- a/drivers/staging/sm750fb/sm750.c
+++ b/drivers/staging/sm750fb/sm750.c
@@ -407,61 +407,29 @@ static inline unsigned int chan_to_field(unsigned int
chan,
return chan << bf->offset;
}
-#ifdef CONFIG_PM
-static int lynxfb_suspend(struct pci_dev *pdev, pm_message_t mesg)
+static int __maybe_unused lynxfb_suspend(struct device *dev)
{
struct fb_info *info;
struct sm750_dev *sm750_dev;
- int ret;
-
- if (mesg.event == pdev->dev.power.power_state.event)
- return 0;
-
- ret = 0;
- sm750_dev = pci_get_drvdata(pdev);
- switch (mesg.event) {
- case PM_EVENT_FREEZE:
- case PM_EVENT_PRETHAW:
- pdev->dev.power.power_state = mesg;
- return 0;
- }
+ sm750_dev = dev_get_drvdata(dev);
console_lock();
- if (mesg.event & PM_EVENT_SLEEP) {
- info = sm750_dev->fbinfo[0];
- if (info)
- /* 1 means do suspend */
- fb_set_suspend(info, 1);
- info = sm750_dev->fbinfo[1];
- if (info)
- /* 1 means do suspend */
- fb_set_suspend(info, 1);
-
- ret = pci_save_state(pdev);
- if (ret) {
- dev_err(&pdev->dev,
- "error:%d occurred in pci_save_state\n", ret);
- goto lynxfb_suspend_err;
- }
-
- ret = pci_set_power_state(pdev, pci_choose_state(pdev, mesg));
- if (ret) {
- dev_err(&pdev->dev,
- "error:%d occurred in pci_set_power_state\n",
- ret);
- goto lynxfb_suspend_err;
- }
- }
-
- pdev->dev.power.power_state = mesg;
+ info = sm750_dev->fbinfo[0];
+ if (info)
+ /* 1 means do suspend */
+ fb_set_suspend(info, 1);
+ info = sm750_dev->fbinfo[1];
+ if (info)
+ /* 1 means do suspend */
+ fb_set_suspend(info, 1);
-lynxfb_suspend_err:
console_unlock();
- return ret;
+ return 0;
}
-static int lynxfb_resume(struct pci_dev *pdev)
+static int __maybe_unused lynxfb_resume(struct device *dev)
{
+ struct pci_dev *pdev = to_pci_dev(dev);
struct fb_info *info;
struct sm750_dev *sm750_dev;
@@ -469,32 +437,10 @@ static int lynxfb_resume(struct pci_dev *pdev)
struct lynxfb_crtc *crtc;
struct lynx_cursor *cursor;
- int ret;
-
- ret = 0;
sm750_dev = pci_get_drvdata(pdev);
console_lock();
- ret = pci_set_power_state(pdev, PCI_D0);
- if (ret) {
- dev_err(&pdev->dev,
- "error:%d occurred in pci_set_power_state\n", ret);
- goto lynxfb_resume_err;
- }
-
- if (pdev->dev.power.power_state.event != PM_EVENT_FREEZE) {
- pci_restore_state(pdev);
- ret = pci_enable_device(pdev);
- if (ret) {
- dev_err(&pdev->dev,
- "error:%d occurred in pci_enable_device\n",
- ret);
- goto lynxfb_resume_err;
- }
- pci_set_master(pdev);
- }
-
hw_sm750_inithw(sm750_dev, pdev);
info = sm750_dev->fbinfo[0];
@@ -523,11 +469,9 @@ static int lynxfb_resume(struct pci_dev *pdev)
pdev->dev.power.power_state.event = PM_EVENT_RESUME;
-lynxfb_resume_err:
console_unlock();
- return ret;
+ return 0;
}
-#endif
static int lynxfb_ops_check_var(struct fb_var_screeninfo *var,
struct fb_info *info)
@@ -1210,15 +1154,14 @@ static const struct pci_device_id smi_pci_table[] = {
MODULE_DEVICE_TABLE(pci, smi_pci_table);
+static SIMPLE_DEV_PM_OPS
Re: [PATCH v1] staging: sm750fb: use generic power management
This patch is compile-tested only Thanks Vaibhav Gupta ___ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
