On Sun, Oct 11, 2020 at 09:35:37AM -0700, Joel Fernandes wrote: > On Fri, Oct 9, 2020 at 4:14 PM Frederic Weisbecker <frederic@xxxxxxxxxx> wrote: > > > > On Wed, Sep 23, 2020 at 11:22:08AM -0400, Joel Fernandes (Google) wrote: > > > Currently, rcu_do_batch() depends on the unsegmented callback list's len field > > > to know how many CBs are executed. This fields counts down from 0 as CBs are > > > dequeued. It is possible that all CBs could not be run because of reaching > > > limits in which case the remaining unexecuted callbacks are requeued in the > > > CPU's segcblist. > > > > > > The number of callbacks that were not requeued are then the negative count (how > > > many CBs were run) stored in the rcl->len which has been counting down on every > > > dequeue. This negative count is then added to the per-cpu segmented callback > > > list's to correct its count. > > > > > > Such a design works against future efforts to track the length of each segment > > > of the segmented callback list. The reason is because > > > rcu_segcblist_extract_done_cbs() will be populating the unsegmented callback > > > list's length field (rcl->len) during extraction. > > > Also, the design of counting down from 0 is confusing and error-prone IMHO. > > > > Right :) > > :) > > > > This commit therefore explicitly counts have many callbacks were executed in > > > > s/have/how > > > > > rcu_do_batch() itself, and uses that to update the per-CPU segcb list's ->len > > > field, without relying on the negativity of rcl->len. > > > > > > Signed-off-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx> > > > > Reviewed-by: Frederic Weisbecker <frederic@xxxxxxxxxx> > > Thanks! Paul would be Ok to make the minor fixup s/have/how/ that > Frederic pointed? But of course! I was waiting until Frederic gets them all reviewed, with an eye to applying and wordsmithing them as a set. > - Joel > (Due to COVID issues at home, I'm intermittently working so advance > apologies for slow replies.) And I hope that this is going as well as it possibly can! Thanx, Paul
