Re: Re: upstart: please update to latest upstream version
Steve Langasek wrote:
> The meme that systemd is better than upstart because it doesn't depend on
> a shell is poppycock. No one has done any benchmarking to support the claim
> that /bin/sh is a bottleneck for upstart (particularly not on Debian or
This misrepresents the systemd position. Not using a shell is faster,
but that's not the only reason to avoid shell. Avoid shell in cases
where it's unnecessary would still be a net win even if it was somewhat
slower.
> OTOH, there are plenty of examples
> of how the limited use of upstart's built-in support for shell scripts makes
> for much more maintainable - and locally-modifiable - startup behavior than
> if this were all implemented in C.
This doesn't seem to match the experience of other people.
You made similar questionable claims earlier (see my reply then at
http://lists.debian.org/debian-devel/2011/07/msg00365.html
). You imply systemd design decisions are based solely on boot speed, to
the detriment of other features. This claim is false. You may disagree
with those design decisions, but it's a fact that they were chosen
because they were (and are) considered good for goals like
maintainability rather than just boot speed. You make claims about
"hard-coding policy into C", but have given no examples of concrete
problems or limitations. What is actually hard-coded so hard that it
would cause problems?
Reply to: