On Wed, 27 Apr 2022, Matthew Wilcox wrote:
On Wed, Apr 27, 2022 at 10:33:31AM -0700, Andrew Morton wrote:
On Wed, 27 Apr 2022 14:08:39 +0000 Liam Howlett <liam.howlett@xxxxxxxxxx> wrote:
> The benchmarks are around the same as they have always been.
So it's presently a wash.
That makes "the plan" (below) really critical, otherwise there seems
little point in merging this code at this time?
Please send me many very soothing words about how confident we should
be that the plan will be implemented and that it shall be good?
Yes, performance-wise it's a wash. However, Davidlohr was very
impressed that it was a wash because we're actually getting rid of three
data structures here; the linked list, the rbtree and the vmacache.
His opinion was that we should push the maple tree in now, in advance
of the future RCU uses.
Yes I like the maple tree, and at this stage I don't think we can ask
for more from this series wrt the MM - albeit there seems to still be
some folks reporting breakage. Fundamentally I see Liam's work to (re)move
complexity out of the MM (not to say that the actual maple tree is not
complex) by consolidating the three complimentary data structures very
much worth it considering performance does not take a hit. This was
very much a turn off with the range locking approach, which worst case
scenario incurred in prohibitive overhead. Also as Liam and Matthew
have mentioned, RCU opens up a lot of nice performance opportunities,
and in addition academia[1] has shown outstanding scalability of address
spaces with the foundation of replacing the locked rbtree with RCU
aware trees.
[1] https://pdos.csail.mit.edu/papers/rcuvm:asplos12.pdf
Thanks,
Davidlohr
