Messages in this thread | | | Date | Wed, 05 Dec 2018 12:25:39 +0100 | From | Roman Penyaev <> | Subject | Re: [RFC PATCH 1/1] epoll: use rwlock in order to reduce ep_poll_callback() contention |
| |
On 2018-12-05 00:59, Andrea Parri wrote: > Hi Roman, > > On Tue, Dec 04, 2018 at 12:50:58PM +0100, Roman Penyaev wrote: >> On 2018-12-03 18:34, Linus Torvalds wrote: > >> > This also ends up making the memory ordering of "xchg()" very very >> > important. Yes, we've documented it as being an ordering op, but I'm >> > not sure we've relied on it this directly before. >> >> Seems exit_mm() does exactly the same, the following chunk: >> >> up_read(&mm->mmap_sem); >> >> self.task = current; >> self.next = xchg(&core_state->dumper.next, &self); >> >> >> At least code pattern looks similar. > > Maybe add a comment on top of (your) xchg() to note/justify these > memory > ordering requirements? As Paul said: "if there are races, this would > help force them to happen" (and simplify the review, this/future).
Hi Andrea,
Sure, this path is tricky, so will I cook something up.
-- Roman
| |