2016-07-10 20:07 GMT+02:00 Denys Vlasenko <vda.li...@googlemail.com>: > On Fri, Jul 8, 2016 at 5:25 PM, Etienne Champetier > <champetier.etie...@gmail.com> wrote: >>> B) To accomplish _what_? You seem to want to use the new API because it >>> exists. >> >> unblock (2 times) faster, nothing more > > I think for now the applet should not be accepted. > > It uses an API which is so new many build environments would not have it. > Which means it should be set to "not selected by default" > in defconfig, which is not typical; or it should be worked around > with using hardcoded syscall numbers (ugly). > > The resulting binary will not work on many deployed kernels > (this is a nuisance: you have your stuff work on your > well-updated desktop distro, but on your router it does not. Grrrr). > > The above could be ignored if the applet would be a really useful > addition. But it is not: dd from /dev/random works almost as well > for the stated purpose.
I agree that the API is a bit new (2 years) so if you use this applet in an LXC, this will work on your updated host but fail on your good old rhel host I will keep this patch in LEDE (OpenWrt) tree for now > > Yes, blocking for 3 minutes rather than 6 minutes is an improvement, > but would 3 minute pause on boot be acceptable for many? I suspect > it's still bad; and if I would have that system at hand, I'd experiment > with making it initialize random pool faster. Basically, > I'd look into kernel source, find what makes it feed entropy pool, > and do that activity a lot in a separate process, while dd is blocked. Just to be sure we are on the same page, I'm not trying to fix a particular router brand or model, I'm trying to improve the whole distro for all supported arch/brand/model (on my free time) _______________________________________________ busybox mailing list busybox@busybox.net http://lists.busybox.net/mailman/listinfo/busybox
