From: Peter Zijlstra <peterz@infradead.org>
To: Oleg Nesterov <oleg@redhat.com>
Cc: mingo@kernel.org, andrii@kernel.org,
linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org,
rostedt@goodmis.org, mhiramat@kernel.org, jolsa@kernel.org,
clm@meta.com, paulmck@kernel.org
Subject: Re: [PATCH v2 11/11] perf/uprobe: Add uretprobe timer
Date: Thu, 11 Jul 2024 18:06:49 +0200 [thread overview]
Message-ID: <20240711160649.GA38270@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20240711155542.GA34311@noisy.programming.kicks-ass.net>
On Thu, Jul 11, 2024 at 05:55:42PM +0200, Peter Zijlstra wrote:
> On Thu, Jul 11, 2024 at 05:00:54PM +0200, Peter Zijlstra wrote:
>
> > Let me ponder that a little, I *can* make it work, but all 'solutions'
> > I've come up with so far are really rather vile.
>
> This is the least horrible solution I could come up with...
>
> ---
> --- a/include/linux/uprobes.h
> +++ b/include/linux/uprobes.h
> @@ -83,6 +83,7 @@ struct uprobe_task {
>
> struct timer_list ri_timer;
> struct callback_head ri_task_work;
> + bool ri_work_pending;
> struct task_struct *task;
> };
>
> --- a/kernel/events/uprobes.c
> +++ b/kernel/events/uprobes.c
> @@ -1797,9 +1797,8 @@ void uprobe_free_utask(struct task_struc
> t->utask = NULL;
> }
>
> -static void return_instance_task_work(struct callback_head *head)
> +static void __return_instance_put(struct uprobe_task *utask)
> {
> - struct uprobe_task *utask = container_of(head, struct uprobe_task, ri_task_work);
> struct return_instance *ri;
>
> for (ri = utask->return_instances; ri; ri = ri->next) {
> @@ -1815,9 +1814,43 @@ static void return_instance_task_work(st
> }
> }
>
> +static void return_instance_task_work(struct callback_head *head)
> +{
> + struct uprobe_task *utask = container_of(head, struct uprobe_task, ri_task_work);
> + utask->ri_work_pending = false;
> + __return_instance_put(utask);
> +}
> +
> +static int return_instance_blocked(struct task_struct *p, void *arg)
> +{
> + unsigned int state = READ_ONCE(p->__state);
> +
> + if (state == TASK_RUNNING || state == TASK_WAKING)
> + return 0;
> +
> + smp_rmb();
> + if (p->on_rq)
> + return 0;
> +
> + /*
> + * Per __task_needs_rq_locked() we now have: !p->on_cpu and only hold
> + * p->pi_lock, and can consiter the task fully blocked.
> + */
> +
> + __return_instance_put(p->utask);
PREEMPT_RT might not like this though, doing the full RI iteration under
a raw_spinlock_t...
I just can't think of anything saner just now. Oh well, let me go make
dinner, perhaps something will come to me.
next prev parent reply other threads:[~2024-07-11 16:06 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-11 11:02 [PATCH v2 00/11] perf/uprobe: Optimize uprobes Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 01/11] perf/uprobe: Re-indent labels Peter Zijlstra
2024-07-11 11:58 ` Jiri Olsa
2024-07-11 12:07 ` Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 02/11] perf/uprobe: Remove spurious whitespace Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 03/11] rbtree: Provide rb_find_rcu() / rb_find_add_rcu() Peter Zijlstra
2024-07-12 20:23 ` Andrii Nakryiko
2024-07-15 11:21 ` Peter Zijlstra
2024-07-15 17:13 ` Andrii Nakryiko
2024-07-11 11:02 ` [PATCH v2 04/11] perf/uprobe: RCU-ify find_uprobe() Peter Zijlstra
2024-07-11 13:59 ` Masami Hiramatsu
2024-07-11 11:02 ` [PATCH v2 05/11] perf/uprobe: Simplify UPROBE_HANDLER_REMOVE logic Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 06/11] perf/uprobe: SRCU-ify uprobe->consumer list Peter Zijlstra
2024-07-12 21:06 ` Andrii Nakryiko
2024-07-15 11:25 ` Peter Zijlstra
2024-07-15 17:30 ` Andrii Nakryiko
2024-07-11 11:02 ` [PATCH v2 07/11] perf/uprobe: Split uprobe_unregister() Peter Zijlstra
2024-07-12 21:10 ` Andrii Nakryiko
2024-07-11 11:02 ` [PATCH v2 08/11] perf/uprobe: Convert (some) uprobe->refcount to SRCU Peter Zijlstra
2024-07-11 14:03 ` Jiri Olsa
2024-07-12 21:21 ` Andrii Nakryiko
2024-07-11 11:02 ` [PATCH v2 09/11] srcu: Add __srcu_clone_read_lock() Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 10/11] perf/uprobe: Convert single-step and uretprobe to SRCU Peter Zijlstra
2024-07-11 16:06 ` Oleg Nesterov
2024-07-11 18:42 ` Peter Zijlstra
2024-07-12 10:26 ` Oleg Nesterov
2024-07-12 21:28 ` Andrii Nakryiko
2024-07-15 11:59 ` Peter Zijlstra
2024-07-11 11:02 ` [PATCH v2 11/11] perf/uprobe: Add uretprobe timer Peter Zijlstra
2024-07-11 13:19 ` Oleg Nesterov
2024-07-11 15:00 ` Peter Zijlstra
2024-07-11 15:55 ` Peter Zijlstra
2024-07-11 16:06 ` Peter Zijlstra [this message]
2024-07-12 21:43 ` Andrii Nakryiko
2024-07-15 11:41 ` Peter Zijlstra
2024-07-15 17:34 ` Andrii Nakryiko
2024-07-12 4:57 ` [PATCH v2 00/11] perf/uprobe: Optimize uprobes Andrii Nakryiko
2024-07-12 9:13 ` Peter Zijlstra
2024-07-12 13:10 ` Peter Zijlstra
2024-07-12 15:29 ` Andrii Nakryiko
2024-07-15 14:45 ` Peter Zijlstra
2024-07-15 17:10 ` Andrii Nakryiko
2024-07-15 18:10 ` Andrii Nakryiko
2024-07-19 18:42 ` Andrii Nakryiko
2024-07-27 0:18 ` Andrii Nakryiko
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=20240711160649.GA38270@noisy.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=andrii@kernel.org \
--cc=clm@meta.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=paulmck@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.