From: Oleg Nesterov <oleg@redhat.com>
To: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>,
linux-trace-kernel@vger.kernel.org, peterz@infradead.org,
rostedt@goodmis.org, mhiramat@kernel.org, bpf@vger.kernel.org,
linux-kernel@vger.kernel.org, jolsa@kernel.org,
paulmck@kernel.org
Subject: Re: [PATCH 7/8] uprobes: perform lockless SRCU-protected uprobes_tree lookup
Date: Thu, 8 Aug 2024 16:29:16 +0200 [thread overview]
Message-ID: <20240808142916.GF8020@redhat.com> (raw)
In-Reply-To: <CAEf4BzYbXzt7RXB962OLEd3xoQcPfT1MFw5JcHSmRzPx-Etm_A@mail.gmail.com>
On 08/07, Andrii Nakryiko wrote:
>
> So, any ideas how we can end up with "corrupted" root on lockless
> lookup with rb_find_rcu()?
I certainly can't help ;) I know ABSOLUTELY NOTHING about rb or any
other tree.
But,
> This seems to be the very first lockless
> RB-tree lookup use case in the tree,
Well, latch_tree_find() is supposed to be rcu-safe afaics, and
__lt_erase() is just rb_erase(). So it is not the 1st use case.
See also the "Notes on lockless lookups" comment in lib/rbtree.c.
So it seems that rb_erase() is supposed to be rcu-safe. However
it uses __rb_change_child(), not __rb_change_child_rcu().
Not that I think this can explain the problem, and on x86
__smp_store_release() is just WRITE_ONCE, but looks confusing...
Oleg.
next prev parent reply other threads:[~2024-08-08 14:29 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-31 21:42 [PATCH 0/8] uprobes: RCU-protected hot path optimizations Andrii Nakryiko
2024-07-31 21:42 ` [PATCH 1/8] rbtree: provide rb_find_rcu() / rb_find_add_rcu() Andrii Nakryiko
2024-07-31 21:42 ` [PATCH 2/8] uprobes: revamp uprobe refcounting and lifetime management Andrii Nakryiko
2024-08-01 11:09 ` Jiri Olsa
2024-08-01 16:49 ` Andrii Nakryiko
2024-08-01 22:07 ` Andrii Nakryiko
2024-08-02 8:50 ` Oleg Nesterov
2024-08-02 14:58 ` Andrii Nakryiko
2024-08-02 22:19 ` Oleg Nesterov
2024-08-02 11:11 ` Jiri Olsa
2024-08-02 15:03 ` Andrii Nakryiko
2024-08-05 13:44 ` Oleg Nesterov
2024-08-05 17:29 ` Andrii Nakryiko
2024-08-06 10:45 ` Oleg Nesterov
2024-07-31 21:42 ` [PATCH 3/8] uprobes: protected uprobe lifetime with SRCU Andrii Nakryiko
2024-08-01 12:23 ` Liao, Chang
2024-08-01 16:49 ` Andrii Nakryiko
2024-08-02 1:30 ` Liao, Chang
2024-08-05 14:51 ` Oleg Nesterov
2024-08-05 17:31 ` Andrii Nakryiko
2024-07-31 21:42 ` [PATCH 4/8] uprobes: get rid of enum uprobe_filter_ctx in uprobe filter callbacks Andrii Nakryiko
2024-07-31 21:42 ` [PATCH 5/8] uprobes: travers uprobe's consumer list locklessly under SRCU protection Andrii Nakryiko
2024-08-01 14:27 ` Jiri Olsa
2024-08-01 16:49 ` Andrii Nakryiko
2024-08-05 15:59 ` Oleg Nesterov
2024-08-05 17:31 ` Andrii Nakryiko
2024-08-06 10:54 ` Oleg Nesterov
2024-07-31 21:42 ` [PATCH 6/8] perf/uprobe: split uprobe_unregister() Andrii Nakryiko
2024-08-02 2:41 ` Liao, Chang
2024-08-02 15:05 ` Andrii Nakryiko
2024-08-05 20:01 ` Andrii Nakryiko
2024-08-06 1:50 ` Liao, Chang
2024-08-07 13:17 ` Oleg Nesterov
2024-08-07 15:24 ` Andrii Nakryiko
2024-07-31 21:42 ` [PATCH 7/8] uprobes: perform lockless SRCU-protected uprobes_tree lookup Andrii Nakryiko
2024-08-07 17:14 ` Andrii Nakryiko
2024-08-08 10:04 ` Oleg Nesterov
2024-08-08 14:29 ` Oleg Nesterov [this message]
2024-08-08 17:00 ` Andrii Nakryiko
2024-08-08 13:40 ` Oleg Nesterov
2024-08-10 14:00 ` Oleg Nesterov
2024-07-31 21:42 ` [PATCH 8/8] uprobes: switch to RCU Tasks Trace flavor for better performance Andrii Nakryiko
2024-08-01 9:35 ` Peter Zijlstra
2024-08-01 16:49 ` Andrii Nakryiko
2024-08-01 18:05 ` Paul E. McKenney
2024-08-07 13:29 ` [PATCH 0/8] uprobes: RCU-protected hot path optimizations Oleg Nesterov
2024-08-07 15:24 ` Andrii Nakryiko
2024-08-07 17:11 ` Oleg Nesterov
2024-08-07 17:31 ` Andrii Nakryiko
2024-08-07 18:24 ` Oleg Nesterov
2024-08-08 7:51 ` Liao, Chang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240808142916.GF8020@redhat.com \
--to=oleg@redhat.com \
--cc=andrii.nakryiko@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).