| Message ID | 20230518074403.2894799-1-m.szyprowski@samsung.com (mailing list archive) |
|---|---|
| State | Accepted |
| Delegated to: | Chanwoo Choi |
| Headers | show |
| Series | PM / devfreq: exynos: add Exynos PPMU as a soft module dependency | expand |
On 18/05/2023 09:44, Marek Szyprowski wrote: > Commit adf8238ef403 ("ARM: dts: exynos: move exynos-bus nodes out of soc > in Exynos4412") changed the order of the exynos-bus nodes, what results > in different probe order of the Exynos Bus devices. Although the driver > properly handles the deferred probe and all devices seems to be finally > properly registered, this change revealed some kind of a bug related to > PPMU counters registration and passive governor operation. Usually in 1 > of 10 boots this results in complete board freeze during loading of the > kernel modules. > > To avoid that freeze, ensure that the Exynos PPMU driver is already > loaded before the Exynos Bus driver starts probing. > > Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Best regards, Krzysztof
On 23. 5. 18. 16:44, Marek Szyprowski wrote: > Commit adf8238ef403 ("ARM: dts: exynos: move exynos-bus nodes out of soc > in Exynos4412") changed the order of the exynos-bus nodes, what results > in different probe order of the Exynos Bus devices. Although the driver > properly handles the deferred probe and all devices seems to be finally > properly registered, this change revealed some kind of a bug related to > PPMU counters registration and passive governor operation. Usually in 1 > of 10 boots this results in complete board freeze during loading of the > kernel modules. > > To avoid that freeze, ensure that the Exynos PPMU driver is already > loaded before the Exynos Bus driver starts probing. > > Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> > --- > This is a follow-up of the following discussion: > https://lore.kernel.org/all/d287ca9f-b056-d39a-aa93-b0e2cb279f73@linaro.org/ > --- > drivers/devfreq/exynos-bus.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c > index 88414445adf3..245898f1a88e 100644 > --- a/drivers/devfreq/exynos-bus.c > +++ b/drivers/devfreq/exynos-bus.c > @@ -518,6 +518,7 @@ static struct platform_driver exynos_bus_platdrv = { > }; > module_platform_driver(exynos_bus_platdrv); > > +MODULE_SOFTDEP("pre: exynos_ppmu"); > MODULE_DESCRIPTION("Generic Exynos Bus frequency driver"); > MODULE_AUTHOR("Chanwoo Choi <cw00.choi@samsung.com>"); > MODULE_LICENSE("GPL v2"); Applied it. Thanks.
diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c index 88414445adf3..245898f1a88e 100644 --- a/drivers/devfreq/exynos-bus.c +++ b/drivers/devfreq/exynos-bus.c @@ -518,6 +518,7 @@ static struct platform_driver exynos_bus_platdrv = { }; module_platform_driver(exynos_bus_platdrv); +MODULE_SOFTDEP("pre: exynos_ppmu"); MODULE_DESCRIPTION("Generic Exynos Bus frequency driver"); MODULE_AUTHOR("Chanwoo Choi <cw00.choi@samsung.com>"); MODULE_LICENSE("GPL v2");
Commit adf8238ef403 ("ARM: dts: exynos: move exynos-bus nodes out of soc in Exynos4412") changed the order of the exynos-bus nodes, what results in different probe order of the Exynos Bus devices. Although the driver properly handles the deferred probe and all devices seems to be finally properly registered, this change revealed some kind of a bug related to PPMU counters registration and passive governor operation. Usually in 1 of 10 boots this results in complete board freeze during loading of the kernel modules. To avoid that freeze, ensure that the Exynos PPMU driver is already loaded before the Exynos Bus driver starts probing. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> --- This is a follow-up of the following discussion: https://lore.kernel.org/all/d287ca9f-b056-d39a-aa93-b0e2cb279f73@linaro.org/ --- drivers/devfreq/exynos-bus.c | 1 + 1 file changed, 1 insertion(+)