On Mon, May 29, 2017 at 10:19 AM, Felix Fietkau <n...@nbd.name> wrote: > On 2017-05-19 09:45, Alexandru Ardelean wrote: >> Admittedly, the semantic is a bit different, in the sense >> that there are no progressive retry timeouts. >> >> ubus_auto_connect() uses 1 second fixed retry intervals. >> Whereas the old logic would start at 50 millisecs and >> progress up to 1 second. >> >> Other than that, the rest should be the same overall logic. >> This should cleanup the code a bit. >> >> For reference, the `ubus_add_uloop()` call is handled >> by ubus_auto_connect() as well. >> >> Signed-off-by: Alexandru Ardelean <ardeleana...@gmail.com> > procd intentionally uses different semantics, since it's responsible for > launching ubus in the first place. > > This patch causes extra delay and extra error messages, so I don't think > we should do something like this at all. > > - Felix
Fine from me. Will close. Thanks :) Alex _______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
