On Wed, Aug 15, 2018 at 07:26:30PM -0400, Kent Overstreet wrote: > Small patch series to > - firstly, refactor generic_file_buffered_read enough that it can be modified > in more interesting ways without going insane, and then > > - secondly, change it to use find_get_pages_contig() to batch up the page > operations, and then copy data to userspace in a separate loop that touches > no other shared cachelines. > > I've been seeing profiles where the radix tree lookups in the buffered read path > are a shockingly large portion of the profile (around 25%, if memory serves) - > that's what this patch series is addressing. I've benchmarked small block reads > as well, performance there is unaffected or slightly improved (it's within the > margin of error). > /me didn't review the patches, but... Could you share how you benchmarked it? Despite the fact I'm curious about it, it's going to be interesting the 'proof' of such improvement. Cheers > And as a bonus, the code that was all in generic_file_buffered_read() is now > _drastically_ easier to follow and modify. I haven't done as much refactoring as > I could have, I kept as much of the structure of the old code as I could just to > make things easier on myself, but I'm still pretty happy with the result. > > Kent Overstreet (2): > fs: Break generic_file_buffered_read up into multiple functions > fs: generic_file_buffered_read() now uses find_get_pages_contig > > mm/filemap.c | 486 +++++++++++++++++++++++++++++---------------------- > 1 file changed, 273 insertions(+), 213 deletions(-) > > -- > 2.18.0 > -- Carlos
