Re: [PATCH 2/2] PM: domain: use per-genpd lockdep class

[Bjorn Andersson]

On Thu 17 Jun 04:07 CDT 2021, Ulf Hansson wrote:

> + Rajendra
> 
> On Tue, 15 Jun 2021 at 17:55, Bjorn Andersson
> <bjorn.andersson@xxxxxxxxxx> wrote:
> >
> > On Tue 15 Jun 05:17 CDT 2021, Ulf Hansson wrote:
> >
> > > + Mark
> > >
> > > On Fri, 11 Jun 2021 at 16:34, Dmitry Baryshkov
> > > <dmitry.baryshkov@xxxxxxxxxx> wrote:
> > > >
> > > > Added Stephen to Cc list
> > > >
> > > > On Fri, 11 Jun 2021 at 16:50, Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
> > > > >
> > > > > On Fri, 11 Jun 2021 at 12:15, Dmitry Baryshkov
> > > > > <dmitry.baryshkov@xxxxxxxxxx> wrote:
> > > > > >
> > > > > > In case of nested genpds it is easy to get the following warning from
> > > > > > lockdep, because all genpd's mutexes share same locking class. Use the
> > > > > > per-genpd locking class to stop lockdep from warning about possible
> > > > > > deadlocks. It is not possible to directly use genpd nested locking, as
> > > > > > it is not the genpd code calling genpd. There are interim calls to
> > > > > > regulator core.
> > > > > >
> > > > > > [    3.030219] ============================================
> > > > > > [    3.030220] WARNING: possible recursive locking detected
> > > > > > [    3.030221] 5.13.0-rc3-00054-gf8f0a2f2b643-dirty #2480 Not tainted
> > > > > > [    3.030222] --------------------------------------------
> > > > > > [    3.030223] kworker/u16:0/7 is trying to acquire lock:
> > > > > > [    3.030224] ffffde0eabd29aa0 (&genpd->mlock){+.+.}-{3:3}, at: genpd_lock_mtx+0x18/0x2c
> > > > > > [    3.030236]
> > > > > > [    3.030236] but task is already holding lock:
> > > > > > [    3.030236] ffffde0eabcfd6d0 (&genpd->mlock){+.+.}-{3:3}, at: genpd_lock_mtx+0x18/0x2c
> > > > > > [    3.030240]
> > > > > > [    3.030240] other info that might help us debug this:
> > > > > > [    3.030240]  Possible unsafe locking scenario:
> > > > > > [    3.030240]
> > > > > > [    3.030241]        CPU0
> > > > > > [    3.030241]        ----
> > > > > > [    3.030242]   lock(&genpd->mlock);
> > > > > > [    3.030243]   lock(&genpd->mlock);
> > > > > > [    3.030244]
> > > > > > [    3.030244]  *** DEADLOCK ***
> > > > > > [    3.030244]
> > > > > > [    3.030244]  May be due to missing lock nesting notation
> > > > > > [    3.030244]
> > > > > > [    3.030245] 6 locks held by kworker/u16:0/7:
> > > > > > [    3.030246]  #0: ffff6cca00010938 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x1f0/0x730
> > > > > > [    3.030252]  #1: ffff8000100c3db0 (deferred_probe_work){+.+.}-{0:0}, at: process_one_work+0x1f0/0x730
> > > > > > [    3.030255]  #2: ffff6cca00ce3188 (&dev->mutex){....}-{3:3}, at: __device_attach+0x3c/0x184
> > > > > > [    3.030260]  #3: ffffde0eabcfd6d0 (&genpd->mlock){+.+.}-{3:3}, at: genpd_lock_mtx+0x18/0x2c
> > > > > > [    3.030264]  #4: ffff8000100c3968 (regulator_ww_class_acquire){+.+.}-{0:0}, at: regulator_lock_dependent+0x6c/0x1b0
> > > > > > [    3.030270]  #5: ffff6cca00a59158 (regulator_ww_class_mutex){+.+.}-{3:3}, at: regulator_lock_recursive+0x94/0x1d0
> > > > > > [    3.030273]
> > > > > > [    3.030273] stack backtrace:
> > > > > > [    3.030275] CPU: 6 PID: 7 Comm: kworker/u16:0 Not tainted 5.13.0-rc3-00054-gf8f0a2f2b643-dirty #2480
> > > > > > [    3.030276] Hardware name: Qualcomm Technologies, Inc. Robotics RB5 (DT)
> > > > > > [    3.030278] Workqueue: events_unbound deferred_probe_work_func
> > > > > > [    3.030280] Call trace:
> > > > > > [    3.030281]  dump_backtrace+0x0/0x1a0
> > > > > > [    3.030284]  show_stack+0x18/0x24
> > > > > > [    3.030286]  dump_stack+0x108/0x188
> > > > > > [    3.030289]  __lock_acquire+0xa20/0x1e0c
> > > > > > [    3.030292]  lock_acquire.part.0+0xc8/0x320
> > > > > > [    3.030294]  lock_acquire+0x68/0x84
> > > > > > [    3.030296]  __mutex_lock+0xa0/0x4f0
> > > > > > [    3.030299]  mutex_lock_nested+0x40/0x50
> > > > > > [    3.030301]  genpd_lock_mtx+0x18/0x2c
> > > > > > [    3.030303]  dev_pm_genpd_set_performance_state+0x94/0x1a0
> > > > > > [    3.030305]  reg_domain_enable+0x28/0x4c
> > > > > > [    3.030308]  _regulator_do_enable+0x420/0x6b0
> > > > > > [    3.030310]  _regulator_enable+0x178/0x1f0
> > > > > > [    3.030312]  regulator_enable+0x3c/0x80
> > > > >
> > > > > At a closer look, I am pretty sure that it's the wrong code design
> > > > > that triggers this problem, rather than that we have a real problem in
> > > > > genpd. To put it simply, the code in genpd isn't designed to work like
> > > > > this. We will end up in circular looking paths, leading to deadlocks,
> > > > > sooner or later if we allow the above code path.
> > > > >
> > > > > To fix it, the regulator here needs to be converted to a proper PM
> > > > > domain. This PM domain should be assigned as the parent to the one
> > > > > that is requested to be powered on.
> > > >
> > > > This more or less resembles original design, replaced per review
> > > > request to use separate regulator
> > > > (https://lore.kernel.org/linux-arm-msm/160269659638.884498.4031967462806977493@xxxxxxxxxxxxxxxxxxxxxxxxxx/,
> > > > https://lore.kernel.org/linux-arm-msm/20201023131925.334864-1-dmitry.baryshkov@xxxxxxxxxx/).
> > >
> > > Thanks for the pointers. In hindsight, it looks like the
> > > "regulator-fixed-domain" DT binding wasn't the right thing.
> > >
> > > Fortunately, it looks like the problem can be quite easily fixed, by
> > > moving to a correct model of the domain hierarchy.
> > >
> >
> > Can you give some pointers to how we actually fix this?
> >
> > The problem that lead us down this path is that drivers/clk/qcom/gdsc.c
> > describes power domains, which are parented by domains provided by
> > drivers/soc/qcom/rpmhpd.c.
> >
> > But I am unable to find a way for the gdsc driver to get hold of the
> > struct generic_pm_domain of the resources exposed by the rpmhpd driver.
> 
> You don't need a handle to the struct generic_pm_domain, to assign a
> parent/child domain. Instead you can use of_genpd_add_subdomain(),
> which takes two "struct of_phandle_args*" corresponding to the
> parent/child device nodes of the genpd providers and then let genpd
> internally do the look up.
> 
> As an example, you may have a look at how the PM domain topology in
> drivers/cpuidle/cpuidle-psci-domain.c are being created.
> 

This seems to do exactly what I was looking for, just different from any
other part of the kernel...

> >
> >
> > The second thing that Dmitry's regulator driver does is to cast the
> > appropriate performance state vote on the rpmhpd resource, but I _think_
> > we can do that using OPP tables in the gdsc client's node...
> 
> Yes, it looks like using an OPP table and to specify a
> "required-opps", at some device node is the right thing to do.
> 
> In this case, I wonder if the "required-opps" belongs to the genpd
> provider node of the new power-domain (as it seems like it only
> supports one fixed performance state when it's powered on). On the
> other hand, specifying this at the consumer node should work as well,
> I think.
> 

I actually think that the single level relates to the needs of the
DISP_CC_MDSS_MDP_CLK clock rate, which we in the DPU node scale further
using an opp table.

So I think it would be appropriate to ensure that we vote on the
performance level from the display driver(s). But this needs some more
investigation.

I don't think enabling MDSS_GDSC requires the performance level
directly.

> Actually, this relates to the changes [1] that Rajendra is working on
> with "assigned-performance-state" (that we agreed to move to
> OPP/required-opps) for genpd.
> 

Might be, but my recent escapades in this area indicates that we do want
to drive the performance state dynamically, and that the current vote is
essentially setting a "minimum".

Regards,
Bjorn

> >
> > > Beyond this, perhaps we should consider removing the
> > > "regulator-fixed-domain" DT property, as to avoid similar problems
> > > from cropping up?
> > >
> >
> > Currently there's a single user upstream, but we have the exact same
> > problem in at least 3-4 platforms with patches in the pipeline.
> >
> > In order to avoid breakage with existing DT I would prefer to see
> > regulator-fixed-domain to live for at least one kernel release beyond
> > the introduction of the other model.
> 
> Yes, this seems reasonable to me.
> 
> As Mark suggested, let's mark the regulator-fixed-domain DT property
> as deprecated and remove it once we have the new solution in place.
> 
> [...]
> 
> Kind regards
> Uffe