On Tue, 9 Jun 2020 16:10:27 +0200 Roland Hieber <rhi@xxxxxxxxxxxxxx> wrote: > First, apply the correct heading level based on the section numbering. > > Then remove leading section numbers altogether, since they can quickly > get out of order when done manually (see the double "5.3" in events.rst, > and the out of order numbering in histogram.rst, where even some > sections are numbered and others are not). Finally, section numbers are > autogenerated anyways in the PDF builds, leading to strange doubly > numerated sections like "14.2.3 6.2 ‘hist’ trigger examples". > > Signed-off-by: Roland Hieber <rhi@xxxxxxxxxxxxxx> So I'm totally in favor of a change like this, but a quick grep turned up a problem: > diff --git a/Documentation/trace/events.rst b/Documentation/trace/events.rst > index f792b1959a33..b6efbf0b17d7 100644 > --- a/Documentation/trace/events.rst > +++ b/Documentation/trace/events.rst [...] > @@ -84,8 +84,8 @@ When reading one of these enable files, there are four results: > - X - there is a mixture of events enabled and disabled > - ? - this file does not affect any event > > -2.3 Boot option > ---------------- > +Boot option > +----------- > > In order to facilitate early boot debugging, use boot option:: > > @@ -94,13 +94,13 @@ In order to facilitate early boot debugging, use boot option:: > event-list is a comma separated list of events. See section 2.1 for event > format. An explicit section number like that no longer works after you've made this change. Such things need to be made into proper references. Thanks, jon
