Re: Improving the performance of system boot by eliding unused rc.d scripts and caching the result

Robert Elz <[email protected]>
Newsgroups gmane.os.netbsd.devel.userlevel
Message-ID <[email protected]>
    Date:        Wed, 19 Aug 2026 23:27:14 -0700
    From:        Jason Thorpe <[email protected]>
    Message-ID:  <[email protected]>

  | Feedback incorporated:
  |
  |     https://www.netbsd.org/~thorpej/rcorder-cache-diff-v4.txt

I see nothing to seriously complain about in that version.   There
is one line that ends
		.... && \
where the \ isn't needed (harms nothing either) - in general any sh
command that ends with an operator which implies the command cannot
possibly end here will just continue reading until there is a place
the command could end (not necessarily where intended, \ is needed when
the command could end at some point, but there is more to come).  There
are also a few places where the syntax doesn't permit a \n, so if one is
needed a \ before it is required, to make the \n vanish as far as the parser
is concerned, one example is between a redirection operator and the word
which follows (the file name or whatever).

&& (or || or | ) cannot possibly be the end, so if one of those ends a
command line, no \ is needed to hide the following \n.   (There are more
cases, but most of those are so obvious that no-one uses a \ anyway.)

(There was at least one other line that ended with \ only to have &&
as the first token on the next line - the \ is needed for that, but
wouldn't be if the && were moved to the previous line, that one was
like that in the original, pre-changes, as well).

These are minor (irrelevant) nits though.

kre
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.