| Message ID | 20240819040310.2801422-1-ruanjinjie@huawei.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [-next] spi: bcm63xx: Fix missing pm_runtime_disable() | expand |
Hi, On Mon, 19 Aug 2024 at 05:55, Jinjie Ruan <ruanjinjie@huawei.com> wrote: > > The pm_runtime_disable() is missing in the remove function, add it to > align with the probe error path. > > Fixes: 2d13f2ff6073 ("spi: bcm63xx-spi: fix pm_runtime") > Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com> > --- > drivers/spi/spi-bcm63xx.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/spi/spi-bcm63xx.c b/drivers/spi/spi-bcm63xx.c > index 289f8a94980b..0531b6f3eef3 100644 > --- a/drivers/spi/spi-bcm63xx.c > +++ b/drivers/spi/spi-bcm63xx.c > @@ -614,6 +614,8 @@ static void bcm63xx_spi_remove(struct platform_device *pdev) > /* reset spi block */ > bcm_spi_writeb(bs, 0, SPI_INT_MASK); > > + pm_runtime_disable(&pdev->dev); > + How about using devm_pm_runtime_enable() instead in the probe path? Then we don't need to call _disable() manually. > /* HW shutdown */ > clk_disable_unprepare(bs->clk); > } > -- > 2.34.1 > Best Regards, Jonas Gorski
On 2024/8/19 15:52, Jonas Gorski wrote: > Hi, > > On Mon, 19 Aug 2024 at 05:55, Jinjie Ruan <ruanjinjie@huawei.com> wrote: >> >> The pm_runtime_disable() is missing in the remove function, add it to >> align with the probe error path. >> >> Fixes: 2d13f2ff6073 ("spi: bcm63xx-spi: fix pm_runtime") >> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com> >> --- >> drivers/spi/spi-bcm63xx.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/drivers/spi/spi-bcm63xx.c b/drivers/spi/spi-bcm63xx.c >> index 289f8a94980b..0531b6f3eef3 100644 >> --- a/drivers/spi/spi-bcm63xx.c >> +++ b/drivers/spi/spi-bcm63xx.c >> @@ -614,6 +614,8 @@ static void bcm63xx_spi_remove(struct platform_device *pdev) >> /* reset spi block */ >> bcm_spi_writeb(bs, 0, SPI_INT_MASK); >> >> + pm_runtime_disable(&pdev->dev); >> + > > How about using devm_pm_runtime_enable() instead in the probe path? > Then we don't need to call _disable() manually. Hi, Jonas I think that is good. I grep the commit log and there is a example: https://lore.kernel.org/all/20240605131533.20037-2-raag.jadav@intel.com/ I'll change it to use devm_pm_runtime_enable() to fix it, thank you! > >> /* HW shutdown */ >> clk_disable_unprepare(bs->clk); >> } >> -- >> 2.34.1 >> > > Best Regards, > Jonas Gorski
diff --git a/drivers/spi/spi-bcm63xx.c b/drivers/spi/spi-bcm63xx.c index 289f8a94980b..0531b6f3eef3 100644 --- a/drivers/spi/spi-bcm63xx.c +++ b/drivers/spi/spi-bcm63xx.c @@ -614,6 +614,8 @@ static void bcm63xx_spi_remove(struct platform_device *pdev) /* reset spi block */ bcm_spi_writeb(bs, 0, SPI_INT_MASK); + pm_runtime_disable(&pdev->dev); + /* HW shutdown */ clk_disable_unprepare(bs->clk); }
The pm_runtime_disable() is missing in the remove function, add it to align with the probe error path. Fixes: 2d13f2ff6073 ("spi: bcm63xx-spi: fix pm_runtime") Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com> --- drivers/spi/spi-bcm63xx.c | 2 ++ 1 file changed, 2 insertions(+)