[Intel-gfx] [PATCH v4] drm/i915/debugfs : PM_REQ and PM_RES registers

[Chris Wilson]

Quoting S, Saichandana (2021-02-05 13:05:20)
> 
> 
> > -----Original Message-----
> > From: Chris Wilson <chris at chris-wilson.co.uk>
> > Sent: Friday, January 15, 2021 6:51 PM
> > To: S, Saichandana <saichandana.s at intel.com>; intel-
> > gfx at lists.freedesktop.org
> > Cc: Nikula, Jani <jani.nikula at intel.com>; S, Saichandana
> > <saichandana.s at intel.com>
> > Subject: Re: [Intel-gfx] [PATCH v4] drm/i915/debugfs : PM_REQ and
> > PM_RES registers
> > 
> > Quoting Saichandana S (2021-01-15 13:01:10)
> > > +static int i915_pm_req_res_info(struct seq_file *m, void *unused) {
> > > +       struct drm_i915_private *i915 = node_to_i915(m->private);
> > > +       struct intel_csr *csr = &i915->csr;
> > > +       u32 DC9_status;
> > > +
> > > +       if (!HAS_CSR(i915))
> > > +               return -ENODEV;
> > > +       if (!csr->dmc_payload)
> > > +               return 0;
> > > +       DC9_status = intel_de_read(i915, PM_RSP_DBG_1) &
> > > + PM_RESP_DC9_READY;
> > 
> > A read without waking the device up? Wait until IGT sees this.
> > [wrap with with_intel_runtime_pm(&i915->runtime_pm, wakeref)] -Chris
> >
> >Using wakeref, doesn't allow i915 to go to suspend state, further blocking display DC9 entry. Thus wakeref IGT test will fail.
> >[S, Saichandana]

And yet you cannot read this register without waking the device up.

Do we conclude that this is a bad test design?
-Chris