[Intel-gfx] [PATCH] drm/i915: Restore user forcewake domains across suspend
Tvrtko Ursulin
tvrtko.ursulin at linux.intel.com
Thu Aug 9 11:46:05 UTC 2018
On 08/08/2018 22:08, Chris Wilson wrote:
> On suspend, we cancel the automatic forcewake and clear all other sources
> of forcewake so the machine can sleep before we do suspend. However, we
> expose the forcewake to userspace (only via debugfs, but nevertheless we
> do) and want to restore that upon resume or else our accounting will be
> off and we may not acquire the forcewake before we use it. So record
> which domains we cleared on suspend and reacquire them early on resume.
>
> v2: Hold the spinlock to appease our sanitychecks
>
> Reported-by: Imre Deak <imre.deak at linux.intel.com>
> Fixes: b8473050805f ("drm/i915: Fix forcewake active domain tracking")
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> Cc: Mika Kuoppala <mika.kuoppala at intel.com>
> Cc: Imre Deak <imre.deak at linux.intel.com>
> ---
> drivers/gpu/drm/i915/intel_uncore.c | 44 ++++++++++---------
> drivers/gpu/drm/i915/intel_uncore.h | 1 +
> drivers/gpu/drm/i915/selftests/intel_uncore.c | 2 +-
> 3 files changed, 26 insertions(+), 21 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
> index 284be151f645..cf40361fe181 100644
> --- a/drivers/gpu/drm/i915/intel_uncore.c
> +++ b/drivers/gpu/drm/i915/intel_uncore.c
> @@ -369,8 +369,8 @@ intel_uncore_fw_release_timer(struct hrtimer *timer)
> }
>
> /* Note callers must have acquired the PUNIT->PMIC bus, before calling this. */
> -static void intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv,
> - bool restore)
> +static unsigned int
> +intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv)
> {
> unsigned long irqflags;
> struct intel_uncore_forcewake_domain *domain;
> @@ -422,20 +422,11 @@ static void intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv,
> dev_priv->uncore.funcs.force_wake_put(dev_priv, fw);
>
> fw_domains_reset(dev_priv, dev_priv->uncore.fw_domains);
> -
> - if (restore) { /* If reset with a user forcewake, try to restore */
> - if (fw)
> - dev_priv->uncore.funcs.force_wake_get(dev_priv, fw);
> -
> - if (IS_GEN6(dev_priv) || IS_GEN7(dev_priv))
> - dev_priv->uncore.fifo_count =
> - fifo_free_entries(dev_priv);
> - }
> -
> - if (!restore)
> - assert_forcewakes_inactive(dev_priv);
> + assert_forcewakes_inactive(dev_priv);
>
> spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
> +
> + return fw; /* track the lost user forcewake domains */
> }
>
> static u64 gen9_edram_size(struct drm_i915_private *dev_priv)
> @@ -544,7 +535,7 @@ check_for_unclaimed_mmio(struct drm_i915_private *dev_priv)
> }
>
> static void __intel_uncore_early_sanitize(struct drm_i915_private *dev_priv,
> - bool restore_forcewake)
> + unsigned int restore_forcewake)
> {
> /* clear out unclaimed reg detection bit */
> if (check_for_unclaimed_mmio(dev_priv))
> @@ -559,7 +550,17 @@ static void __intel_uncore_early_sanitize(struct drm_i915_private *dev_priv,
> }
>
> iosf_mbi_punit_acquire();
> - intel_uncore_forcewake_reset(dev_priv, restore_forcewake);
> + intel_uncore_forcewake_reset(dev_priv);
> + if (restore_forcewake) {
> + spin_lock_irq(&dev_priv->uncore.lock);
> + dev_priv->uncore.funcs.force_wake_get(dev_priv,
> + restore_forcewake);
> +
> + if (IS_GEN6(dev_priv) || IS_GEN7(dev_priv))
> + dev_priv->uncore.fifo_count =
> + fifo_free_entries(dev_priv);
> + spin_unlock_irq(&dev_priv->uncore.lock);
> + }
> iosf_mbi_punit_release();
> }
>
> @@ -568,13 +569,16 @@ void intel_uncore_suspend(struct drm_i915_private *dev_priv)
> iosf_mbi_punit_acquire();
> iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
> &dev_priv->uncore.pmic_bus_access_nb);
> - intel_uncore_forcewake_reset(dev_priv, false);
> + dev_priv->uncore.fw_domains_user =
> + intel_uncore_forcewake_reset(dev_priv);
> iosf_mbi_punit_release();
> }
>
> void intel_uncore_resume_early(struct drm_i915_private *dev_priv)
> {
> - __intel_uncore_early_sanitize(dev_priv, true);
> + __intel_uncore_early_sanitize(dev_priv,
> + dev_priv->uncore.fw_domains_user);
> +
> iosf_mbi_register_pmic_bus_access_notifier(
> &dev_priv->uncore.pmic_bus_access_nb);
> i915_check_and_clear_faults(dev_priv);
> @@ -1555,7 +1559,7 @@ void intel_uncore_init(struct drm_i915_private *dev_priv)
>
> intel_uncore_edram_detect(dev_priv);
> intel_uncore_fw_domains_init(dev_priv);
> - __intel_uncore_early_sanitize(dev_priv, false);
> + __intel_uncore_early_sanitize(dev_priv, 0);
I wonder if a tweak to not have a parameter but function acts on
uncore.fw_domains_user on it's own, and consumes it after restoring form
it. But it is just a different tweak on the same theme so never mind.
>
> dev_priv->uncore.unclaimed_mmio_check = 1;
> dev_priv->uncore.pmic_bus_access_nb.notifier_call =
> @@ -1642,7 +1646,7 @@ void intel_uncore_fini(struct drm_i915_private *dev_priv)
> iosf_mbi_punit_acquire();
> iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
> &dev_priv->uncore.pmic_bus_access_nb);
> - intel_uncore_forcewake_reset(dev_priv, false);
> + intel_uncore_forcewake_reset(dev_priv);
> iosf_mbi_punit_release();
> }
>
> diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h
> index 2fbe93178fb2..137d8ac856fe 100644
> --- a/drivers/gpu/drm/i915/intel_uncore.h
> +++ b/drivers/gpu/drm/i915/intel_uncore.h
> @@ -104,6 +104,7 @@ struct intel_uncore {
>
> enum forcewake_domains fw_domains;
> enum forcewake_domains fw_domains_active;
> + enum forcewake_domains fw_domains_user;
Maybe call it fw_domains_saved so it is obvious it is special since user
domains are not always tracked in it.
>
> u32 fw_set;
> u32 fw_clear;
> diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
> index 47bc5b2ddb56..81d9d31042a9 100644
> --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
> +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
> @@ -160,7 +160,7 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri
> i915_reg_t reg = { offset };
>
> iosf_mbi_punit_acquire();
> - intel_uncore_forcewake_reset(dev_priv, false);
> + intel_uncore_forcewake_reset(dev_priv);
> iosf_mbi_punit_release();
>
> check_for_unclaimed_mmio(dev_priv);
>
AFAICS it was obviously broken and this fixes it.
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
Regards,
Tvrtko
More information about the Intel-gfx
mailing list