Received-SPF: pass (google.com: domain of linux-kernel+bounces-208023-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
From: Thomas Gleixner <tglx@linutronix.de>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>, Sam Sun <samsun1006219@gmail.com>,
 linux-kernel@vger.kernel.org, x86@kernel.org,
 syzkaller-bugs@googlegroups.com, jpoimboe@kernel.org, jbaron@akamai.com,
 ardb@kernel.org, mingo@redhat.com, Borislav Petkov <bp@alien8.de>,
 dave.hansen@linux.intel.com, hpa@zytor.com, xrivendell7@gmail.com, Greg
 Kroah-Hartman <gregkh@linuxfoundation.org>, Tejun Heo <tj@kernel.org>
Subject: Re: [Linux kernel bug] WARNING in static_key_slow_inc_cpuslocked
In-Reply-To: <20240610064651.GS8774@noisy.programming.kicks-ass.net>
References: <20240609090431.3af238bc@gandalf.local.home>
 <87o78axlbm.ffs@tglx> <20240609102530.0a292b07@rorschach.local.home>
 <87le3exfx2.ffs@tglx> <87h6e2xdg1.ffs@tglx>
 <20240610064651.GS8774@noisy.programming.kicks-ass.net>
Date: Mon, 10 Jun 2024 12:34:14 +0200
Message-ID: <87bk49xf15.ffs@tglx>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain

On Mon, Jun 10 2024 at 08:46, Peter Zijlstra wrote:
> On Sun, Jun 09, 2024 at 06:56:14PM +0200, Thomas Gleixner wrote:
>
>> Ok. Now I found if for real. It's in the jump label core:
>> 
>> CPU0                            CPU1
>> 
>> static_key_slow_dec()
>>  static_key_slow_try_dec()
>> 
>>    key->enabled == 1
>>    val = atomic_fetch_add_unless(&key->enabled, -1, 1);
>>    if (val == 1)
>>    	return false;
>> 
>>    jump_label_lock();
>>    if (atomic_dec_and_test(&key->enabled)) {
>>       --> key->enabled == 0
>>       __jump_label_update()
>> 
>>                                 static_key_slow_dec()
>>                                  static_key_slow_try_dec()
>> 
>>                                     key->enabled == 0
>>                                     val = atomic_fetch_add_unless(&key->enabled, -1, 1);
>> 
>>                                     --> key->enabled == -1 <- FAIL
>> 
>> static_key_slow_try_dec() is buggy. It needs similar logic as
>> static_key_slow_try_inc() to work correctly.
>> 
>> It's not only the 0, key->enabled can be -1 when the other CPU is in the
>> slow path of enabling it.
>
> Well, the -1 thing is in the 0->1 path, that is, the very first enabler.
>
> That *should* not race with a disabler. If it does, there is external
> confusion. (As I think the follow up email shows..)

Right, but all of this is too fragile. Let me send out those patches.