On Tue, 25 Feb 2014, Dave Hansen wrote: > On 02/25/2014 03:09 PM, Alexander Graf wrote: > >> Couldn't we also (maybe in parallel) just teach the sysctl userspace > >> about sysfs? This way we don't have to do parallel sysctls and sysfs > >> for *EVERYTHING* in the kernel: > >> > >> sysfs.kernel.mm.transparent_hugepage.enabled=enabled > > > > It's pretty hard to filter this. We definitely do not want to expose all of sysfs through /proc/sys. But how do we know which files are actual configuration and which ones are dynamic system introspection data? > > > > We could add a filter, but then we can just as well stick with the manual approach I followed here :). > > Maybe not stick it under /proc/sys, but teach sysctl(8) about them. I > guess at the moment, sysctl says that it's tied to /proc/sys: > > > DESCRIPTION > > sysctl is used to modify kernel parameters at runtime. The parameters available are those listed under /proc/sys/. Procfs is required > > for sysctl support in Linux. You can use sysctl to both read and write sysctl data. > > But surely that's not set in stone just because the manpage says so. :) What I still don't get is why you need this? My distribution (Debian) has a sysfsutils package which provides a /etc/sysfs.conf / /etc/sysfs.d/foo exactly like /etc/sysctl.conf. Don't other distributions have something like this? c'ya sven-haegar -- Three may keep a secret, if two of them are dead. - Ben F. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>