Re: [PATCH 00/10] perf/uprobe: Optimize uprobes

[Peter Zijlstra <peterz@infradead.org>]

On Tue, Jul 09, 2024 at 07:56:51AM +0900, Masami Hiramatsu wrote:
> On Mon, 08 Jul 2024 11:12:41 +0200
> Peter Zijlstra <peterz@infradead.org> wrote:
> 
> > Hi!
> > 
> > These patches implement the (S)RCU based proposal to optimize uprobes.
> > 
> > On my c^Htrusty old IVB-EP -- where each (of the 40) CPU calls 'func' in a
> > tight loop:
> > 
> >   perf probe -x ./uprobes test=func
> >   perf stat -ae probe_uprobe:test  -- sleep 1
> > 
> >   perf probe -x ./uprobes test=func%return
> >   perf stat -ae probe_uprobe:test__return -- sleep 1
> > 
> > PRE:
> > 
> >   4,038,804      probe_uprobe:test
> >   2,356,275      probe_uprobe:test__return
> > 
> > POST:
> > 
> >   7,216,579      probe_uprobe:test
> >   6,744,786      probe_uprobe:test__return
> > 
> 
> Good results! So this is another series of Andrii's batch register?

Yeah, it is my counter proposal. I didn't much like the refcounting
thing he ended up with, and his own numbers show the refcounting remains
a significant problem.

These patches mostly do away with the refcounting entirely -- except for
the extremely rare case where you let a return probe sit for over a
second without anything else happening.