Message-ID: <20230913132404.wepr7l7o5qegy7hw@revolver>
Date:   Wed, 13 Sep 2023 09:24:04 -0400
From:   "Liam R. Howlett" <Liam.Howlett@...cle.com>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     paulmck@...nel.org, Andrew Morton <akpm@...ux-foundation.org>,
        maple-tree@...ts.infradead.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        linux-renesas-soc@...r.kernel.org,
        Shanker Donthineni <sdonthineni@...dia.com>
Subject: Re: [PATCH v2 1/2] maple_tree: Disable mas_wr_append() when other
 readers are possible

* Geert Uytterhoeven <geert@...ux-m68k.org> [230913 09:15]:
> Hi Paul,
> 
> On Tue, Sep 12, 2023 at 12:00 PM Paul E. McKenney <paulmck@...nel.org> wrote:
> > On Tue, Sep 12, 2023 at 10:34:44AM +0200, Geert Uytterhoeven wrote:
> > > On Tue, Sep 12, 2023 at 10:30 AM Paul E. McKenney <paulmck@...nel.org> wrote:
> > > > On Tue, Sep 12, 2023 at 10:23:37AM +0200, Geert Uytterhoeven wrote:
> > > > > On Tue, Sep 12, 2023 at 10:14 AM Paul E. McKenney <paulmck@...nel.org> wrote:
> > > > > > On Mon, Sep 11, 2023 at 07:54:52PM -0400, Liam R. Howlett wrote:
> > > > > > > * Paul E. McKenney <paulmck@...nel.org> [230906 14:03]:
> > > > > > > > On Wed, Sep 06, 2023 at 01:29:54PM -0400, Liam R. Howlett wrote:
> > > > > > > > > * Paul E. McKenney <paulmck@...nel.org> [230906 13:24]:
> > > > > > > > > > On Wed, Sep 06, 2023 at 11:23:25AM -0400, Liam R. Howlett wrote:
> > > > > > > > > > > (Adding Paul & Shanker to Cc list.. please see below for why)
> > > > > > > > > > >
> > > > > > > > > > > Apologies on the late response, I was away and have been struggling to
> > > > > > > > > > > get a working PPC32 test environment.
> > > > > > > > > > >
> > > > > > > > > > > * Geert Uytterhoeven <geert@...ux-m68k.org> [230829 12:42]:
> > > > > > > > > > > >     Hi Liam,
> > > > > > > > > > > >
> > > > > > > > > > > > On Fri, 18 Aug 2023, Liam R. Howlett wrote:
> > > > > > > > > > > > > The current implementation of append may cause duplicate data and/or
> > > > > > > > > > > > > incorrect ranges to be returned to a reader during an update.  Although
> > > > > > > > > > > > > this has not been reported or seen, disable the append write operation
> > > > > > > > > > > > > while the tree is in rcu mode out of an abundance of caution.
> > > > > > > > > > >
> > > > > > > > > > > ...
> > > > > > > > > > > > >
> > > > > > >
> > > > > > > ...
> > > > > > >
> > > > > > > > > > > > RCU-related configs:
> > > > > > > > > > > >
> > > > > > > > > > > >     $ grep RCU .config
> > > > > > > > > > > >     # RCU Subsystem
> > > > > > > > > > > >     CONFIG_TINY_RCU=y
> > > > > >
> > > > > > I must have been asleep last time I looked at this.  I was looking at
> > > > > > Tree RCU.  Please accept my apologies for my lapse.  :-/
> > > > > >
> > > > > > However, Tiny RCU's call_rcu() also avoids enabling IRQs, so I would
> > > > > > have said the same thing, albeit after looking at a lot less RCU code.
> > > > > >
> > > > > > TL;DR:
> > > > > >
> > > > > > 1.      Try making the __setup_irq() function's call to mutex_lock()
> > > > > >         instead be as follows:
> > > > > >
> > > > > >         if (!mutex_trylock(&desc->request_mutex))
> > > > > >                 mutex_lock(&desc->request_mutex);
> > > > > >
> > > > > >         This might fail if __setup_irq() has other dependencies on a
> > > > > >         fully operational scheduler.
> > > > > >
> > > > > > 2.      Move that ppc32 call to __setup_irq() much later, most definitely
> > > > > >         after interrupts have been enabled and the scheduler is fully
> > > > > >         operational.  Invoking mutex_lock() before that time is not a
> > > > > >         good idea.  ;-)
> > > > >
> > > > > There is no call to __setup_irq() from arch/powerpc/?
> > > >
> > > > Glad it is not just me, given that I didn't see a direct call, either.  So
> > > > later in this email, I asked Liam to put a WARN_ON_ONCE(irqs_disabled())
> > > > just before that mutex_lock() in __setup_irq().
> > > >
> > > > Either way, invoking mutex_lock() early in boot before interrupts have
> > > > been enabled is a bad idea.  ;-)
> > >
> > > I'll add that WARN_ON_ONCE() too, and will report back later today...
> >
> > Thank you, looking forward to hearing the outcome!
> 
> On a typical arm32 system with global or architectured timer, I get:
> 
> WARNING: CPU: 0 PID: 0 at kernel/irq/manage.c:1587 __setup_irq+0xcc/0x6d0
> CPU: 0 PID: 0 Comm: swapper/0 Not tainted
> 6.6.0-rc1-shmobile-02354-g24e058b77f5a #1655
> Hardware name: Generic R-Car Gen2 (Flattened Device Tree)
>  unwind_backtrace from show_stack+0x10/0x14
>  show_stack from dump_stack_lvl+0x40/0x4c
>  dump_stack_lvl from __warn+0x78/0x10c
>  __warn from warn_slowpath_fmt+0x90/0x11c
>  warn_slowpath_fmt from __setup_irq+0xcc/0x6d0
>  __setup_irq from __request_percpu_irq+0xb8/0xd0
>  __request_percpu_irq from set_smp_ipi_range+0x88/0xdc
>  set_smp_ipi_range from gic_of_init+0x1a4/0x4d8
>  gic_of_init from of_irq_init+0x1f0/0x320
>  of_irq_init from init_IRQ+0x74/0x104
>  init_IRQ from start_kernel+0x360/0x5d0
>  start_kernel from 0x0
> 
> Likewise on arm64 with architectured timer.
> But on these systems I do not see the issue I reported.
> 
> On RISC-V:
> 
> ------------[ cut here ]------------
> WARNING: CPU: 0 PID: 0 at kernel/irq/manage.c:1587 __setup_irq+0x4e6/0x5ee
> Modules linked in:
> CPU: 0 PID: 0 Comm: swapper Not tainted 6.6.0-rc1-02355-g63165363c6a3 #63
> Hardware name: Renesas SMARC EVK based on r9a07g043f01 (DT)
> epc : __setup_irq+0x4e6/0x5ee
>  ra : __setup_irq+0x38/0x5ee
> epc : ffffffff80049472 ra : ffffffff80048fc4 sp : ffffffff81203cd0
>  gp : ffffffff812ee760 tp : ffffffff8120d5c0 t0 : ffffffd801854300
>  t1 : 0000000000000000 t2 : ffffffff81000ad8 s0 : ffffffff81203d20
>  s1 : ffffffd801855000 a0 : 0000000000000001 a1 : ffffffd801855000
>  a2 : ffffffd801854280 a3 : 0000000000000001 a4 : 0000000000000000
>  a5 : 0000000000000000 a6 : ffffffd801852108 a7 : ffffffd801852100
>  s2 : ffffffd801854280 s3 : 0000000000000005 s4 : ffffffff812c54c0
>  s5 : 0000000000000005 s6 : ffffffff80dd83a0 s7 : ffffffff805c0cc0
>  s8 : ffffffd801855018 s9 : 0000000000000000 s10: 0000000000000000
>  s11: 000000007bf638a0 t3 : 0000000000000000 t4 : 0000000000000002
>  t5 : ffffffff812882a0 t6 : 0000000000000001
> status: 0000000200000100 badaddr: 0000000000000000 cause: 0000000000000003
> [<ffffffff80049472>] __setup_irq+0x4e6/0x5ee
> [<ffffffff800497a8>] __request_percpu_irq+0x98/0xcc
> [<ffffffff8082501e>] riscv_timer_init_dt+0x186/0x22e
> [<ffffffff80824b62>] timer_probe+0x62/0xd2
> [<ffffffff80803c36>] time_init+0x86/0xa6
> [<ffffffff80800ae2>] start_kernel+0x436/0x618
> ---[ end trace 0000000000000000 ]---
> 
> Also, no issue here.
> 
> On the affected systems (RZ/A1 and RZ/A2), the WARN_ON_ONCE() did
> not trigger, until I applied Liam's patch ("init/main: Clear boot task
> idle flag"), which got rid of the "Interrupts were enabled early" warning,
> and now tells me, as expected:
> 
> ------------[ cut here ]------------
> WARNING: CPU: 0 PID: 0 at kernel/irq/manage.c:1587 __setup_irq+0xc8/0x654
> CPU: 0 PID: 0 Comm: swapper Not tainted
> 6.6.0-rc1-rskrza1-02357-g237e09fd64b8-dirty #829
> Hardware name: Generic R7S72100 (Flattened Device Tree)
>  unwind_backtrace from show_stack+0x10/0x14
>  show_stack from dump_stack_lvl+0x24/0x3c
>  dump_stack_lvl from __warn+0x74/0xb8
>  __warn from warn_slowpath_fmt+0x78/0xb0
>  warn_slowpath_fmt from __setup_irq+0xc8/0x654
>  __setup_irq from request_threaded_irq+0xac/0x13c
>  request_threaded_irq from timer_of_init+0x238/0x2c8
>  timer_of_init from ostm_init+0x98/0x208
>  ostm_init from timer_probe+0x90/0xe4
>  timer_probe from start_kernel+0x2c0/0x488
>  start_kernel from 0x0
> ---[ end trace 0000000000000000 ]---
> 
> However, Liam's patch causes lots of warnings on the other systems...
> 

What patch?  The "init/main" patch?

What systems and what are they?

Thanks,
Liam

Powered by blists - more mailing lists