From: Hans de Goede <j.w.r.degoede@gmail.com>
To: "Daniel Vetter" <daniel.vetter@intel.com>,
"Jani Nikula" <jani.nikula@linux.intel.com>,
"Ville Syrjälä" <ville.syrjala@linux.intel.com>,
"Imre Deak" <imre.deak@intel.com>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Ingo Molnar" <mingo@redhat.com>,
"H . Peter Anvin" <hpa@zytor.com>
Cc: Hans de Goede <hdegoede@redhat.com>,
intel-gfx <intel-gfx@lists.freedesktop.org>,
dri-devel@lists.freedesktop.org, x86@kernel.org,
linux-kernel@vger.kernel.org
Subject: [PATCH v5 2/2] drm/i915: Acquire PUNIT->PMIC bus for intel_uncore_forcewake_reset()
Date: Thu, 19 Oct 2017 13:16:20 +0200 [thread overview]
Message-ID: <20171019111620.26761-3-hdegoede@redhat.com> (raw)
In-Reply-To: <20171019111620.26761-1-hdegoede@redhat.com>
intel_uncore_forcewake_reset() does forcewake puts and gets as such
we need to make sure that no-one tries to access the PUNIT->PMIC bus
(on systems where this bus is shared) while it runs, otherwise bad
things happen.
Normally this is taken care of by the i915_pmic_bus_access_notifier()
which does an intel_uncore_forcewake_get(FORCEWAKE_ALL) when some other
driver tries to access the PMIC bus, so that later forcewake gets are
no-ops (for the duration of the bus access).
But intel_uncore_forcewake_reset gets called in 3 cases:
1) Before registering the pmic_bus_access_notifier
2) After unregistering the pmic_bus_access_notifier
3) To reset forcewake state on a GPU reset
In all 3 cases the i915_pmic_bus_access_notifier() protection is
insufficient.
This commit fixes this race by calling iosf_mbi_punit_acquire() before
calling intel_uncore_forcewake_reset(). In the case where it is called
directly after unregistering the pmic_bus_access_notifier, we need to
hold the punit-lock over both calls to avoid a race where
intel_uncore_fw_release_timer() may execute between the 2 calls.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
---
Changes in v2:
-Rebase on current (July 6th 2017) drm-next
Changes in v3:
-Keep punit acquired / locked over the unregister + forcewake_reset
call combo to avoid a race hitting between the 2 calls
-This requires modifying iosf_mbi_unregister_pmic_bus_access_notifier
to not take the lock itself, since we are the only users this is done
in this same commit
Changes in v4:
-Fix missing rename in doc-comment
-Add and use iosf_mbi_assert_punit_acquired() helper
-Add missing acquire surrounding intel_uncore_forcewake_reset() inside
intel_uncore_check_forcewake_domains()
-Add Imre's Reviewed-by
Changes in v5:
-Separate out arch/x86 iosf_mbi changes into a separate patch
---
drivers/gpu/drm/i915/intel_uncore.c | 17 +++++++++++++----
drivers/gpu/drm/i915/selftests/intel_uncore.c | 3 +++
2 files changed, 16 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 8c2ce81f01c2..0da81faf3981 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -229,6 +229,7 @@ intel_uncore_fw_release_timer(struct hrtimer *timer)
return HRTIMER_NORESTART;
}
+/* 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)
{
@@ -237,6 +238,8 @@ static void intel_uncore_forcewake_reset(struct drm_i915_private *dev_priv,
int retry_count = 100;
enum forcewake_domains fw, active_domains;
+ iosf_mbi_assert_punit_acquired();
+
/* Hold uncore.lock across reset to prevent any register access
* with forcewake not set correctly. Wait until all pending
* timers are run before holding.
@@ -416,14 +419,18 @@ static void __intel_uncore_early_sanitize(struct drm_i915_private *dev_priv,
GT_FIFO_CTL_RC6_POLICY_STALL);
}
+ iosf_mbi_punit_acquire();
intel_uncore_forcewake_reset(dev_priv, restore_forcewake);
+ iosf_mbi_punit_release();
}
void intel_uncore_suspend(struct drm_i915_private *dev_priv)
{
- iosf_mbi_unregister_pmic_bus_access_notifier(
+ 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);
+ iosf_mbi_punit_release();
}
void intel_uncore_resume_early(struct drm_i915_private *dev_priv)
@@ -1315,12 +1322,14 @@ void intel_uncore_init(struct drm_i915_private *dev_priv)
void intel_uncore_fini(struct drm_i915_private *dev_priv)
{
- iosf_mbi_unregister_pmic_bus_access_notifier(
- &dev_priv->uncore.pmic_bus_access_nb);
-
/* Paranoia: make sure we have disabled everything before we exit. */
intel_uncore_sanitize(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);
+ iosf_mbi_punit_release();
}
static const struct reg_whitelist {
diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
index 3cac22eb47ce..733d87fe7737 100644
--- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
@@ -148,7 +148,10 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri
for_each_set_bit(offset, valid, FW_RANGE) {
i915_reg_t reg = { offset };
+ iosf_mbi_punit_acquire();
intel_uncore_forcewake_reset(dev_priv, false);
+ iosf_mbi_punit_release();
+
check_for_unclaimed_mmio(dev_priv);
(void)I915_READ(reg);
--
2.14.2
next prev parent reply other threads:[~2017-10-19 11:16 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-19 11:16 [PATCH v5 1/2] x86 / i915 iosf_mbi / PMIC bus access fixes Hans de Goede
2017-10-19 11:16 ` [PATCH v5 1/2] x86/platform/intel/iosf_mbi: Add unlocked PMIC bus access notifier unregister Hans de Goede
2017-10-19 11:16 ` Hans de Goede [this message]
2017-10-31 9:50 ` [PATCH v5 2/2] drm/i915: Acquire PUNIT->PMIC bus for intel_uncore_forcewake_reset() Ingo Molnar
2017-10-31 10:04 ` [Intel-gfx] " Daniel Vetter
2017-10-31 11:54 ` Ingo Molnar
2017-10-31 10:10 ` Hans de Goede
2017-10-30 13:11 ` [PATCH v5 1/2] x86 / i915 iosf_mbi / PMIC bus access fixes Hans de Goede
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=20171019111620.26761-3-hdegoede@redhat.com \
--to=j.w.r.degoede@gmail.com \
--cc=daniel.vetter@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=hdegoede@redhat.com \
--cc=hpa@zytor.com \
--cc=imre.deak@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jani.nikula@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=ville.syrjala@linux.intel.com \
--cc=x86@kernel.org \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®