Re: [PATCH] uprobes: reduce contention on uprobes_tree access

[Masami Hiramatsu (Google) <mhiramat@kernel.org>]

On Fri, 29 Mar 2024 10:33:57 -0700
Andrii Nakryiko <andrii.nakryiko@gmail.com> wrote:

> On Wed, Mar 27, 2024 at 5:45 PM Andrii Nakryiko
> <andrii.nakryiko@gmail.com> wrote:
> >
> > On Wed, Mar 27, 2024 at 5:18 PM Masami Hiramatsu <mhiramat@kernel.org> wrote:
> > >
> > > On Wed, 27 Mar 2024 17:06:01 +0000
> > > Jonthan Haslam <jonathan.haslam@gmail.com> wrote:
> > >
> > > > > > Masami,
> > > > > >
> > > > > > Given the discussion around per-cpu rw semaphore and need for
> > > > > > (internal) batched attachment API for uprobes, do you think you can
> > > > > > apply this patch as is for now? We can then gain initial improvements
> > > > > > in scalability that are also easy to backport, and Jonathan will work
> > > > > > on a more complete solution based on per-cpu RW semaphore, as
> > > > > > suggested by Ingo.
> > > > >
> > > > > Yeah, it is interesting to use per-cpu rw semaphore on uprobe.
> > > > > I would like to wait for the next version.
> > > >
> > > > My initial tests show a nice improvement on the over RW spinlocks but
> > > > significant regression in acquiring a write lock. I've got a few days
> > > > vacation over Easter but I'll aim to get some more formalised results out
> > > > to the thread toward the end of next week.
> > >
> > > As far as the write lock is only on the cold path, I think you can choose
> > > per-cpu RW semaphore. Since it does not do busy wait, the total system
> > > performance impact will be small.
> >
> > No, Masami, unfortunately it's not as simple. In BPF we have BPF
> > multi-uprobe, which can be used to attach to thousands of user
> > functions. It currently creates one uprobe at a time, as we don't
> > really have a batched API. If each such uprobe registration will now
> > take a (relatively) long time, when multiplied by number of attach-to
> > user functions, it will be a horrible regression in terms of
> > attachment/detachment performance.

Ah, got it. So attachment/detachment performance should be counted.

> >
> > So when we switch to per-CPU rw semaphore, we'll need to provide an
> > internal batch uprobe attach/detach API to make sure that attaching to
> > multiple uprobes is still fast.

Yeah, we need such interface like register_uprobes(...).

> >
> > Which is why I was asking to land this patch as is, as it relieves the
> > scalability pains in production and is easy to backport to old
> > kernels. And then we can work on batched APIs and switch to per-CPU rw
> > semaphore.

OK, then I'll push this to for-next at this moment.
Please share if you have a good idea for the batch interface which can be
backported. I guess it should involve updating userspace changes too.

Thank you!

> >
> > So I hope you can reconsider and accept improvements in this patch,
> > while Jonathan will keep working on even better final solution.
> > Thanks!
> >
> > > I look forward to your formalized results :)
> > >
> 
> BTW, as part of BPF selftests, we have a multi-attach test for uprobes
> and USDTs, reporting attach/detach timings:
> $ sudo ./test_progs -v -t uprobe_multi_test/bench
> bpf_testmod.ko is already unloaded.
> Loading bpf_testmod.ko...
> Successfully loaded bpf_testmod.ko.
> test_bench_attach_uprobe:PASS:uprobe_multi_bench__open_and_load 0 nsec
> test_bench_attach_uprobe:PASS:uprobe_multi_bench__attach 0 nsec
> test_bench_attach_uprobe:PASS:uprobes_count 0 nsec
> test_bench_attach_uprobe: attached in   0.120s
> test_bench_attach_uprobe: detached in   0.092s
> #400/5   uprobe_multi_test/bench_uprobe:OK
> test_bench_attach_usdt:PASS:uprobe_multi__open 0 nsec
> test_bench_attach_usdt:PASS:bpf_program__attach_usdt 0 nsec
> test_bench_attach_usdt:PASS:usdt_count 0 nsec
> test_bench_attach_usdt: attached in   0.124s
> test_bench_attach_usdt: detached in   0.064s
> #400/6   uprobe_multi_test/bench_usdt:OK
> #400     uprobe_multi_test:OK
> Summary: 1/2 PASSED, 0 SKIPPED, 0 FAILED
> Successfully unloaded bpf_testmod.ko.
> 
> So it should be easy for Jonathan to validate his changes with this.
> 
> > > Thank you,
> > >
> > > >
> > > > Jon.
> > > >
> > > > >
> > > > > Thank you,
> > > > >
> > > > > >
> > > > > > >
> > > > > > > BTW, how did you measure the overhead? I think spinlock overhead
> > > > > > > will depend on how much lock contention happens.
> > > > > > >
> > > > > > > Thank you,
> > > > > > >
> > > > > > > >
> > > > > > > > [0] https://docs.kernel.org/locking/spinlocks.html
> > > > > > > >
> > > > > > > > Signed-off-by: Jonathan Haslam <jonathan.haslam@gmail.com>
> > > > > > > > ---
> > > > > > > >  kernel/events/uprobes.c | 22 +++++++++++-----------
> > > > > > > >  1 file changed, 11 insertions(+), 11 deletions(-)
> > > > > > > >
> > > > > > > > diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
> > > > > > > > index 929e98c62965..42bf9b6e8bc0 100644
> > > > > > > > --- a/kernel/events/uprobes.c
> > > > > > > > +++ b/kernel/events/uprobes.c
> > > > > > > > @@ -39,7 +39,7 @@ static struct rb_root uprobes_tree = RB_ROOT;
> > > > > > > >   */
> > > > > > > >  #define no_uprobe_events()   RB_EMPTY_ROOT(&uprobes_tree)
> > > > > > > >
> > > > > > > > -static DEFINE_SPINLOCK(uprobes_treelock);    /* serialize rbtree access */
> > > > > > > > +static DEFINE_RWLOCK(uprobes_treelock);      /* serialize rbtree access */
> > > > > > > >
> > > > > > > >  #define UPROBES_HASH_SZ      13
> > > > > > > >  /* serialize uprobe->pending_list */
> > > > > > > > @@ -669,9 +669,9 @@ static struct uprobe *find_uprobe(struct inode *inode, loff_t offset)
> > > > > > > >  {
> > > > > > > >       struct uprobe *uprobe;
> > > > > > > >
> > > > > > > > -     spin_lock(&uprobes_treelock);
> > > > > > > > +     read_lock(&uprobes_treelock);
> > > > > > > >       uprobe = __find_uprobe(inode, offset);
> > > > > > > > -     spin_unlock(&uprobes_treelock);
> > > > > > > > +     read_unlock(&uprobes_treelock);
> > > > > > > >
> > > > > > > >       return uprobe;
> > > > > > > >  }
> > > > > > > > @@ -701,9 +701,9 @@ static struct uprobe *insert_uprobe(struct uprobe *uprobe)
> > > > > > > >  {
> > > > > > > >       struct uprobe *u;
> > > > > > > >
> > > > > > > > -     spin_lock(&uprobes_treelock);
> > > > > > > > +     write_lock(&uprobes_treelock);
> > > > > > > >       u = __insert_uprobe(uprobe);
> > > > > > > > -     spin_unlock(&uprobes_treelock);
> > > > > > > > +     write_unlock(&uprobes_treelock);
> > > > > > > >
> > > > > > > >       return u;
> > > > > > > >  }
> > > > > > > > @@ -935,9 +935,9 @@ static void delete_uprobe(struct uprobe *uprobe)
> > > > > > > >       if (WARN_ON(!uprobe_is_active(uprobe)))
> > > > > > > >               return;
> > > > > > > >
> > > > > > > > -     spin_lock(&uprobes_treelock);
> > > > > > > > +     write_lock(&uprobes_treelock);
> > > > > > > >       rb_erase(&uprobe->rb_node, &uprobes_tree);
> > > > > > > > -     spin_unlock(&uprobes_treelock);
> > > > > > > > +     write_unlock(&uprobes_treelock);
> > > > > > > >       RB_CLEAR_NODE(&uprobe->rb_node); /* for uprobe_is_active() */
> > > > > > > >       put_uprobe(uprobe);
> > > > > > > >  }
> > > > > > > > @@ -1298,7 +1298,7 @@ static void build_probe_list(struct inode *inode,
> > > > > > > >       min = vaddr_to_offset(vma, start);
> > > > > > > >       max = min + (end - start) - 1;
> > > > > > > >
> > > > > > > > -     spin_lock(&uprobes_treelock);
> > > > > > > > +     read_lock(&uprobes_treelock);
> > > > > > > >       n = find_node_in_range(inode, min, max);
> > > > > > > >       if (n) {
> > > > > > > >               for (t = n; t; t = rb_prev(t)) {
> > > > > > > > @@ -1316,7 +1316,7 @@ static void build_probe_list(struct inode *inode,
> > > > > > > >                       get_uprobe(u);
> > > > > > > >               }
> > > > > > > >       }
> > > > > > > > -     spin_unlock(&uprobes_treelock);
> > > > > > > > +     read_unlock(&uprobes_treelock);
> > > > > > > >  }
> > > > > > > >
> > > > > > > >  /* @vma contains reference counter, not the probed instruction. */
> > > > > > > > @@ -1407,9 +1407,9 @@ vma_has_uprobes(struct vm_area_struct *vma, unsigned long start, unsigned long e
> > > > > > > >       min = vaddr_to_offset(vma, start);
> > > > > > > >       max = min + (end - start) - 1;
> > > > > > > >
> > > > > > > > -     spin_lock(&uprobes_treelock);
> > > > > > > > +     read_lock(&uprobes_treelock);
> > > > > > > >       n = find_node_in_range(inode, min, max);
> > > > > > > > -     spin_unlock(&uprobes_treelock);
> > > > > > > > +     read_unlock(&uprobes_treelock);
> > > > > > > >
> > > > > > > >       return !!n;
> > > > > > > >  }
> > > > > > > > --
> > > > > > > > 2.43.0
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Masami Hiramatsu (Google) <mhiramat@kernel.org>
> > > > >
> > > > >
> > > > > --
> > > > > Masami Hiramatsu (Google) <mhiramat@kernel.org>
> > >
> > >
> > > --
> > > Masami Hiramatsu (Google) <mhiramat@kernel.org>


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>