On Fri, 25 Sep 2020 17:14:42 +1000 NeilBrown <neilb@xxxxxxx> wrote: > There are behavioural requirements on the seq_file next() function in > terms of how it updates *pos at end-of-file, and these are now enforced > by a warning. > > I was recently attempting to justify the reason this was needed, and > couldn't remember the details, and didn't find them in the > documentation. > > So I re-read the code until I understood it again, and updated the > documentation to match. > > I also enhanced the text about SEQ_START_TOKEN as it seemed potentially > misleading. > > Cc: Vasily Averin <vvs@xxxxxxxxxxxxx> > Signed-off-by: NeilBrown <neilb@xxxxxxx> > --- > Documentation/filesystems/seq_file.rst | 20 +++++++++++++++++++- > 1 file changed, 19 insertions(+), 1 deletion(-) Applied, thanks. jon
