With the support for loading of backends as modules (see for example:
"staging: zcache: enable zcache to be built/loaded as a module"), the
frontswap_enabled is always set to true ("mm: frontswap: lazy
initialization to allow tmem backends to build/run as modules").The next patch "frontswap: Use static_key instead of frontswap_enabled and frontswap_ops" is are going to convert the frontswap_enabled to be a bit more selective and be on/off depending on whether the backend has registered - and not whether the frontswap API is enabled. The two functions: frontswap_init and frontswap_invalidate_area can be called anytime - they queue up which of the swap devices are active and can use the frontswap API - once the backend is loaded. As such there is no need to check for 'frontswap_enabled' at all. Signed-off-by: Konrad Rzeszutek Wilk <[email protected]> --- include/linux/frontswap.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/include/linux/frontswap.h b/include/linux/frontswap.h index d4f2987..140323b 100644 --- a/include/linux/frontswap.h +++ b/include/linux/frontswap.h @@ -116,14 +116,12 @@ static inline void frontswap_invalidate_page(unsigned type, pgoff_t offset) static inline void frontswap_invalidate_area(unsigned type) { - if (frontswap_enabled) - __frontswap_invalidate_area(type); + __frontswap_invalidate_area(type); } static inline void frontswap_init(unsigned type) { - if (frontswap_enabled) - __frontswap_init(type); + __frontswap_init(type); } #endif /* _LINUX_FRONTSWAP_H */ -- 1.8.0.2 _______________________________________________ devel mailing list [email protected] http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
