All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: paulmck <paulmck@linux.ibm.com>, Will Deacon <will@kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	"Joel Fernandes, Google" <joel@joelfernandes.org>,
	Alan Stern <stern@rowland.harvard.edu>,
	rostedt <rostedt@goodmis.org>,
	Valentin Schneider <valentin.schneider@arm.com>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Boqun Feng <boqun.feng@gmail.com>,
	Will Deacon <will.deacon@arm.com>,
	David Howells <dhowells@redhat.com>
Subject: Re: [PATCH 1/1] Fix: trace sched switch start/stop racy updates
Date: Wed, 21 Aug 2019 11:48:43 -0400 (EDT)	[thread overview]
Message-ID: <52600272.1909.1566402523680.JavaMail.zimbra@efficios.com> (raw)
In-Reply-To: <20190821153325.GD2349@hirez.programming.kicks-ass.net>

----- On Aug 21, 2019, at 8:33 AM, Peter Zijlstra peterz@infradead.org wrote:

> On Wed, Aug 21, 2019 at 06:23:10AM -0700, Paul E. McKenney wrote:
>> On Wed, Aug 21, 2019 at 11:32:01AM +0100, Will Deacon wrote:
> 
>> > and so it is using a store-pair instruction to reduce the complexity in
>> > the immediate generation. Thus, the 64-bit store will only have 32-bit
>> > atomicity. In fact, this is scary because if I change bar to:
>> > 
>> > void bar(u64 *x)
>> > {
>> > 	*(volatile u64 *)x = 0xabcdef10abcdef10;
>> > }
>> > 
>> > then I get:
>> > 
>> > bar:
>> > 	mov	w1, 61200
>> > 	movk	w1, 0xabcd, lsl 16
>> > 	str	w1, [x0]
>> > 	str	w1, [x0, 4]
>> > 	ret
>> > 
>> > so I'm not sure that WRITE_ONCE would even help :/
>> 
>> Well, I can have the LWN article cite your email, then.  So thank you
>> very much!
>> 
>> Is generation of this code for a 64-bit volatile store considered a bug?
>> Or does ARMv8 exclude the possibility of 64-bit MMIO registers?  And I
>> would guess that Thomas and Linus would ask a similar bugginess question
>> for normal stores.  ;-)
> 
> I'm calling this a compiler bug; the way I understand volatile this is
> very much against the intentended use case. That is, this is buggy even
> on UP vs signals or MMIO.

And here is a simpler reproducer on my gcc-8.3.0 (aarch64) compiled with O2:

volatile unsigned long a;
 
void fct(void)
{
        a = 0x1234567812345678ULL;
}

void fct(void)
{
        a = 0x1234567812345678ULL;
   0:   90000000        adrp    x0, 8 <fct+0x8>
   4:   528acf01        mov     w1, #0x5678                     // #22136
   8:   72a24681        movk    w1, #0x1234, lsl #16
   c:   f9400000        ldr     x0, [x0]
  10:   b9000001        str     w1, [x0]
  14:   b9000401        str     w1, [x0, #4]
}
  18:   d65f03c0        ret

And the non-volatile case uses stp (is it a single store to memory ?):

unsigned long a;
  
void fct(void)
{
        a = 0x1234567812345678ULL;
}

void fct(void)
{
        a = 0x1234567812345678ULL;
   0:   90000000        adrp    x0, 8 <fct+0x8>
   4:   528acf01        mov     w1, #0x5678                     // #22136
   8:   72a24681        movk    w1, #0x1234, lsl #16
   c:   f9400000        ldr     x0, [x0]
  10:   29000401        stp     w1, w1, [x0]
}
  14:   d65f03c0        ret

It would probably be a good idea to audit other architectures, since this
is done by the compiler backend.

Thanks,

Mathieu








-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

  reply	other threads:[~2019-08-21 15:48 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-18 10:29 WARNING in tracepoint_probe_register_prio (3) syzbot
2019-08-16  0:11 ` syzbot
2019-08-16 14:26   ` [PATCH 1/1] Fix: trace sched switch start/stop racy updates Mathieu Desnoyers
2019-08-16 16:25     ` Steven Rostedt
2019-08-16 16:48       ` Valentin Schneider
2019-08-16 17:04         ` Steven Rostedt
2019-08-16 17:41           ` Mathieu Desnoyers
2019-08-16 19:18             ` Steven Rostedt
2019-08-16 19:19             ` Alan Stern
2019-08-16 20:44               ` Joel Fernandes
2019-08-16 20:49                 ` Thomas Gleixner
2019-08-16 20:57                   ` Joel Fernandes
2019-08-16 22:27                     ` Valentin Schneider
2019-08-16 22:57                       ` Linus Torvalds
2019-08-17  1:41                         ` Mathieu Desnoyers
2019-08-17  4:52                         ` Paul E. McKenney
2019-08-17  8:28                           ` Linus Torvalds
2019-08-17  8:44                             ` Linus Torvalds
2019-08-17 15:02                               ` Mathieu Desnoyers
2019-08-17 20:03                                 ` Valentin Schneider
2019-08-17 23:00                                   ` Paul E. McKenney
2019-08-19 10:34                                     ` Valentin Schneider
2019-08-17 22:28                             ` Paul E. McKenney
2019-08-20 14:01                           ` Peter Zijlstra
2019-08-20 20:31                             ` Paul E. McKenney
2019-08-20 20:39                               ` Peter Zijlstra
2019-08-20 20:52                                 ` Paul E. McKenney
2019-08-16 21:04                   ` Linus Torvalds
2019-08-17  1:36                     ` Mathieu Desnoyers
2019-08-17  2:13                       ` Steven Rostedt
2019-08-17 14:40                         ` Mathieu Desnoyers
2019-08-17 15:26                           ` Steven Rostedt
2019-08-17 15:55                             ` Mathieu Desnoyers
2019-08-17 16:40                               ` Steven Rostedt
2019-08-17 22:06                                 ` Paul E. McKenney
2019-08-17  8:08                       ` Linus Torvalds
2019-08-20 13:56                         ` Peter Zijlstra
2019-08-20 20:29                           ` Paul E. McKenney
2019-08-21 10:32                             ` Will Deacon
2019-08-21 13:23                               ` Paul E. McKenney
2019-08-21 13:32                                 ` Will Deacon
2019-08-21 13:56                                   ` Paul E. McKenney
2019-08-21 16:22                                     ` Will Deacon
2019-08-21 15:33                                 ` Peter Zijlstra
2019-08-21 15:48                                   ` Mathieu Desnoyers [this message]
2019-08-21 16:14                                     ` Paul E. McKenney
2019-08-21 19:03                                     ` Joel Fernandes
2019-09-09  6:21                           ` Herbert Xu
2019-08-16 20:49                 ` Steven Rostedt
2019-08-16 20:59                   ` Joel Fernandes
2019-08-17  1:25                   ` Mathieu Desnoyers
2019-08-18  9:15                   ` stable markup was " Pavel Machek
2019-08-16 17:19       ` Mathieu Desnoyers
2019-08-16 19:15         ` Steven Rostedt
2019-08-17 14:27           ` Mathieu Desnoyers
2019-08-17 15:42             ` Steven Rostedt
2019-08-17 15:53               ` Mathieu Desnoyers
2019-08-17 16:43                 ` Steven Rostedt
2019-08-16 12:32 ` WARNING in tracepoint_probe_register_prio (3) syzbot
2019-08-16 12:41   ` Sebastian Andrzej Siewior

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=52600272.1909.1566402523680.JavaMail.zimbra@efficios.com \
    --to=mathieu.desnoyers@efficios.com \
    --cc=boqun.feng@gmail.com \
    --cc=dhowells@redhat.com \
    --cc=joel@joelfernandes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulmck@linux.ibm.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=stern@rowland.harvard.edu \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=valentin.schneider@arm.com \
    --cc=will.deacon@arm.com \
    --cc=will@kernel.org \
    --subject='Re: [PATCH 1/1] Fix: trace sched switch start/stop racy updates' \
    /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

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.