| Message ID | 20200625142435.50371e2f@ipc1.ka-ro |
|---|---|
| State | Awaiting Upstream |
| Delegated to: | David Miller |
| Headers | show |
| Series | [BUGFIX] can: m_can: make m_can driver work with sleep state pinconfig | expand |
Lothar On 6/25/20 7:24 AM, Lothar Waßmann wrote: > Hi, > > When trying to use the m_can driver on an stm32mp15 based system, I > found that I could not send or receive any data. > Analyzing the pinctrl registers revealed, that the pins were > configured for sleep state even when the can interfaces were in use. > > Looking at the m_can_platform.c driver I found that: > > commit f524f829b75a ("can: m_can: Create a m_can platform framework") > > introduced a call to m_can_class_suspend() in the m_can_runtime_suspend() > function which wasn't there in the original code and which causes the > pins used by the controller to be configured for sleep state. > > commit 0704c5743694 ("can: m_can_platform: remove unnecessary m_can_class_resume() call") > already removed a bogus call to m_can_class_resume() from the > m_can_runtime_resume() function, but failed to remove the matching > call to m_can_class_suspend() from the m_can_runtime_suspend() function. > > Removing the bogus call to m_can_class_suspend() in the > m_can_runtime_suspend() function fixes this. Thank you for the patch Richard G has already submitted a similar patch https://lore.kernel.org/patchwork/patch/1253401/ Dan
diff --git a/drivers/net/can/m_can/m_can_platform.c b/drivers/net/can/m_can/m_can_platform.c index 38ea5e600fb8..e6d0cb9ee02f 100644 --- a/drivers/net/can/m_can/m_can_platform.c +++ b/drivers/net/can/m_can/m_can_platform.c @@ -144,8 +144,6 @@ static int __maybe_unused m_can_runtime_suspend(struct device *dev) struct net_device *ndev = dev_get_drvdata(dev); struct m_can_classdev *mcan_class = netdev_priv(ndev); - m_can_class_suspend(dev); - clk_disable_unprepare(mcan_class->cclk); clk_disable_unprepare(mcan_class->hclk);
Hi, When trying to use the m_can driver on an stm32mp15 based system, I found that I could not send or receive any data. Analyzing the pinctrl registers revealed, that the pins were configured for sleep state even when the can interfaces were in use. Looking at the m_can_platform.c driver I found that: commit f524f829b75a ("can: m_can: Create a m_can platform framework") introduced a call to m_can_class_suspend() in the m_can_runtime_suspend() function which wasn't there in the original code and which causes the pins used by the controller to be configured for sleep state. commit 0704c5743694 ("can: m_can_platform: remove unnecessary m_can_class_resume() call") already removed a bogus call to m_can_class_resume() from the m_can_runtime_resume() function, but failed to remove the matching call to m_can_class_suspend() from the m_can_runtime_suspend() function. Removing the bogus call to m_can_class_suspend() in the m_can_runtime_suspend() function fixes this. Fixes: f524f829b75a ("can: m_can: Create a m_can platform framework") Fixes: 0704c5743694 ("can: m_can_platform: remove unnecessary m_can_class_resume() call") Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de> --- drivers/net/can/m_can/m_can_platform.c | 2 -- 1 file changed, 2 deletions(-)