Skip to contentSkip to conversation

drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

PATCH In review Sagar Arun Kamble 4 participants 14 messages Nov 5, 2017 – Nov 8, 2017

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/5 drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

Releasing GuC doorbell involves stopping the doorbell unit snooping and executing doorbell deallocation flow in GuC firmware. When GuC is in sane state, both steps will be necessary to release doorbell.
If GuC is hung (possibly leading to i915 reset), only doorbell unit snooping is to be stopped. destroy_doorbell will be called in suspend and reset flows differently in upcoming patches.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_guc_submission.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)
drivers/gpu/drm/i915/i915_guc_submission.c +15 -8
index d14c134..b6486dc 100644
@@ -207,7 +207,7 @@ static int __create_doorbell(struct i915_guc_client *client)
 	return err;
 }
 
-static int __destroy_doorbell(struct i915_guc_client *client)
+static int __destroy_doorbell(struct i915_guc_client *client, bool notify_guc)
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(client->guc);
 	struct guc_doorbell_info *doorbell;
@@ -225,7 +225,10 @@ static int __destroy_doorbell(struct i915_guc_client *client)
 	if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID), 10))
 		WARN_ONCE(true, "Doorbell never became invalid after disable\n");
 
-	return __guc_deallocate_doorbell(client->guc, client->stage_id);
+	if (notify_guc)
+		return __guc_deallocate_doorbell(client->guc, client->stage_id);
+
+	return 0;
 }
 
 static int create_doorbell(struct i915_guc_client *client)
@@ -250,7 +253,11 @@ static int create_doorbell(struct i915_guc_client *client)
 	return ret;
 }
 
-static int destroy_doorbell(struct i915_guc_client *client)
+/*
+ * This function deactivates doorbell monitoring through doorbell unit and
+ * optionally notifies GuC to deallocate the doorbell.
+ */
+static int destroy_doorbell(struct i915_guc_client *client, bool notify_guc)
 {
 	int err;
 
@@ -259,7 +266,7 @@ static int destroy_doorbell(struct i915_guc_client *client)
 	/* XXX: wait for any interrupts */
 	/* XXX: wait for workqueue to drain */
 
-	err = __destroy_doorbell(client);
+	err = __destroy_doorbell(client, notify_guc);
 	if (err)
 		return err;
 
@@ -873,7 +880,7 @@ static int __reset_doorbell(struct i915_guc_client* client, u16 db_id)
 	__update_doorbell_desc(client, db_id);
 	err = __create_doorbell(client);
 	if (!err)
-		err = __destroy_doorbell(client);
+		err = __destroy_doorbell(client, true);
 
 	return err;
 }
@@ -899,7 +906,7 @@ static int guc_init_doorbell_hw(struct intel_guc *guc)
 
 		if (has_doorbell(client)) {
 			/* Borrow execbuf_client (we will recreate it later) */
-			destroy_doorbell(client);
+			destroy_doorbell(client, true);
 			recreate_first_client = true;
 		}
 
@@ -925,7 +932,7 @@ static int guc_init_doorbell_hw(struct intel_guc *guc)
 
 	ret = __create_doorbell(guc->preempt_client);
 	if (ret) {
-		__destroy_doorbell(guc->execbuf_client);
+		__destroy_doorbell(guc->execbuf_client, true);
 		return ret;
 	}
 
@@ -1043,7 +1050,7 @@ static void guc_client_free(struct i915_guc_client *client)
 	/* FIXME: in many cases, by the time we get here the GuC has been
 	 * reset, so we cannot destroy the doorbell properly. Ignore the
 	 * error message for now */
-	destroy_doorbell(client);
+	destroy_doorbell(client, true);
 	guc_stage_desc_fini(client->guc, client);
 	i915_gem_object_unpin_map(client->vma->obj);
 	i915_vma_unpin_and_release(&client->vma);
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

PATCH 2/5 drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Before GT device suspend, i915 should release GuC client doorbells by stopping doorbell controller snooping and doorbell deallocation through GuC. They need to be acquired again on resume. This will also resolve the driver unload issue with GuC, where doorbell deallocation was being done post GEM suspend.
Release function is called from guc_suspend prior to sending sleep action during runtime and drm suspend. Acquiral is different in runtime and drm resume paths as on drm resume we are currently doing full reinit. Release should be done in synchronization with acquire hence GuC suspend/resume along with doorbell release/acquire should be under struct_mutex. Upcoming suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_drv.c            |  3 +++
 drivers/gpu/drm/i915/i915_gem.c            |  5 +++--
 drivers/gpu/drm/i915/i915_guc_submission.c | 20 ++++++++++++++++----
 drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++
 drivers/gpu/drm/i915/intel_guc.c           |  2 ++
 5 files changed, 26 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +3 -0
index e7e9e06..3df8a7d 100644
@@ -47,6 +47,7 @@
 #include <drm/i915_drm.h>
 
 #include "i915_drv.h"
+#include "i915_guc_submission.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
 #include "intel_drv.h"
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct device *kdev)
 
 	intel_guc_resume(dev_priv);
 
+	i915_guc_clients_acquire_doorbells(&dev_priv->guc);
+
 	if (IS_GEN9_LP(dev_priv)) {
 		bxt_disable_dc9(dev_priv);
 		bxt_display_core_init(dev_priv, true);
drivers/gpu/drm/i915/i915_gem.c +3 -2
index 6a71805..cbce714 100644
@@ -29,13 +29,14 @@
 #include <drm/drm_vma_manager.h>
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
+#include "i915_gemfs.h"
 #include "i915_gem_clflush.h"
-#include "i915_vgpu.h"
+#include "i915_guc_submission.h"
 #include "i915_trace.h"
+#include "i915_vgpu.h"
 #include "intel_drv.h"
 #include "intel_frontbuffer.h"
 #include "intel_mocs.h"
-#include "i915_gemfs.h"
 #include <linux/dma-fence-array.h>
 #include <linux/kthread.h>
 #include <linux/reservation.h>
drivers/gpu/drm/i915/i915_guc_submission.c +16 -4
index b6486dc..b989edf 100644
@@ -1047,10 +1047,6 @@ static void guc_client_free(struct i915_guc_client *client)
 	 * Be sure to drop any locks
 	 */
 
-	/* FIXME: in many cases, by the time we get here the GuC has been
-	 * reset, so we cannot destroy the doorbell properly. Ignore the
-	 * error message for now */
-	destroy_doorbell(client, true);
 	guc_stage_desc_fini(client->guc, client);
 	i915_gem_object_unpin_map(client->vma->obj);
 	i915_vma_unpin_and_release(&client->vma);
@@ -1102,6 +1098,21 @@ static void guc_clients_destroy(struct intel_guc *guc)
 	guc_client_free(client);
 }
 
+void i915_guc_clients_acquire_doorbells(struct intel_guc *guc)
+{
+	if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS)
+		return;
+
+	create_doorbell(guc->execbuf_client);
+	create_doorbell(guc->preempt_client);
+}
+
+void i915_guc_clients_release_doorbells(struct intel_guc *guc)
+{
+	destroy_doorbell(guc->preempt_client, true);
+	destroy_doorbell(guc->execbuf_client, true);
+}
+
 static void guc_policy_init(struct guc_policy *policy)
 {
 	policy->execution_quantum = POLICY_DEFAULT_EXECUTION_QUANTUM_US;
@@ -1423,6 +1434,7 @@ int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
 	} else {
 		guc_reset_wq(guc->execbuf_client);
 		guc_reset_wq(guc->preempt_client);
+		i915_guc_clients_acquire_doorbells(guc);
 	}
 
 	err = intel_guc_sample_forcewake(guc);
drivers/gpu/drm/i915/i915_guc_submission.h +2 -0
index cb4353b..1d5cf22 100644
@@ -72,6 +72,8 @@ struct i915_guc_client {
 	u64 submissions[I915_NUM_ENGINES];
 };
 
+void i915_guc_clients_acquire_doorbells(struct intel_guc *guc);
+void i915_guc_clients_release_doorbells(struct intel_guc *guc);
 int i915_guc_submission_init(struct drm_i915_private *dev_priv);
 int i915_guc_submission_enable(struct drm_i915_private *dev_priv);
 void i915_guc_submission_disable(struct drm_i915_private *dev_priv);
drivers/gpu/drm/i915/intel_guc.c +2 -0
index 9678630..b6eb5ac 100644
@@ -274,6 +274,8 @@ int intel_guc_suspend(struct drm_i915_private *dev_priv)
 	if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS)
 		return 0;
 
+	i915_guc_clients_release_doorbells(&dev_priv->guc);
+
 	gen9_disable_guc_interrupts(dev_priv);
 
 	data[0] = INTEL_GUC_ACTION_ENTER_S_STATE;
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw

Re: [PATCH 2/5] drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Quoting Sagar Arun Kamble (2017-11-05 13:39:40)

Quoted 40 lines

Before GT device suspend, i915 should release GuC client doorbells by stopping doorbell controller snooping and doorbell deallocation through GuC. They need to be acquired again on resume. This will also resolve the driver unload issue with GuC, where doorbell deallocation was being done post GEM suspend.
Release function is called from guc_suspend prior to sending sleep action during runtime and drm suspend. Acquiral is different in runtime and drm resume paths as on drm resume we are currently doing full reinit. Release should be done in synchronization with acquire hence GuC suspend/resume along with doorbell release/acquire should be under struct_mutex. Upcoming suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_drv.c            |  3 +++
 drivers/gpu/drm/i915/i915_gem.c            |  5 +++--
 drivers/gpu/drm/i915/i915_guc_submission.c | 20 ++++++++++++++++----
 drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++
 drivers/gpu/drm/i915/intel_guc.c           |  2 ++
 5 files changed, 26 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -0
index e7e9e06..3df8a7d 100644
@@ -47,6 +47,7 @@
 #include <drm/i915_drm.h>
 
 #include "i915_drv.h"
+#include "i915_guc_submission.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
 #include "intel_drv.h"

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct device *kdev) · line 2619

Quoted 4 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct device *kdev)
 
        intel_guc_resume(dev_priv);
 
+       i915_guc_clients_acquire_doorbells(&dev_priv->guc);

intel_guc_acquire_doorbells();

Though, if we need to specialise between resume and runtime_resume, I suggest adding intel_guc_resume() and intel_guc_runtime_resume() entry points. (We probably should find a better way to distinguish those two paths, system_resume vs runtime_resume?)
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw

Re: [PATCH 2/5] drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Quoted 42 lines · Chris Wilson

Quoting Sagar Arun Kamble (2017-11-05 13:39:40)

Quoted 41 lines

Before GT device suspend, i915 should release GuC client doorbells by stopping doorbell controller snooping and doorbell deallocation through GuC. They need to be acquired again on resume. This will also resolve the driver unload issue with GuC, where doorbell deallocation was being done post GEM suspend.
Release function is called from guc_suspend prior to sending sleep action during runtime and drm suspend. Acquiral is different in runtime and drm resume paths as on drm resume we are currently doing full reinit. Release should be done in synchronization with acquire hence GuC suspend/resume along with doorbell release/acquire should be under struct_mutex. Upcoming suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
  drivers/gpu/drm/i915/i915_drv.c            |  3 +++
  drivers/gpu/drm/i915/i915_gem.c            |  5 +++--
  drivers/gpu/drm/i915/i915_guc_submission.c | 20 ++++++++++++++++----
  drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++
  drivers/gpu/drm/i915/intel_guc.c           |  2 ++
  5 files changed, 26 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -0
index e7e9e06..3df8a7d 100644
@@ -47,6 +47,7 @@
  #include <drm/i915_drm.h>
 
  #include "i915_drv.h"
+#include "i915_guc_submission.h"
  #include "i915_trace.h"
  #include "i915_vgpu.h"
  #include "intel_drv.h"
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct device *kdev)
 
Quoted 4 lines

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct device *kdev) · line 2619

Quoted 3 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
         intel_guc_resume(dev_priv);
 
+       i915_guc_clients_acquire_doorbells(&dev_priv->guc);

intel_guc_acquire_doorbells();

Prefixed "i915_guc_clients" since this modifies submission state (clients/doorbells). Should have kept dev_priv as parameter. what should be the correct option here: intel_guc*(guc) or i915_guc*(dev_priv)

Quoted 5 lines

Though, if we need to specialise between resume and runtime_resume, I suggest adding intel_guc_resume() and intel_guc_runtime_resume() entry points. (We probably should find a better way to distinguish those two paths, system_resume vs runtime_resume?)

Yes. these will be added in the upcoming GuC suspend/resume restructuring series. Functionality along runtime/system suspend path will be same for GuC. Will defer in the resume path since we are doing full gem reinit on resume from system suspend.

Quoted 1 line

-Chris

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#4 lore raw

Re: [PATCH 2/5] drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Quoting Sagar Arun Kamble (2017-11-07 06:05:01)

Quoted 45 lines · Chris Wilson
Quoted 42 lines

Quoting Sagar Arun Kamble (2017-11-05 13:39:40)

Quoted 41 lines

Before GT device suspend, i915 should release GuC client doorbells by stopping doorbell controller snooping and doorbell deallocation through GuC. They need to be acquired again on resume. This will also resolve the driver unload issue with GuC, where doorbell deallocation was being done post GEM suspend.
Release function is called from guc_suspend prior to sending sleep action during runtime and drm suspend. Acquiral is different in runtime and drm resume paths as on drm resume we are currently doing full reinit. Release should be done in synchronization with acquire hence GuC suspend/resume along with doorbell release/acquire should be under struct_mutex. Upcoming suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
  drivers/gpu/drm/i915/i915_drv.c            |  3 +++
  drivers/gpu/drm/i915/i915_gem.c            |  5 +++--
  drivers/gpu/drm/i915/i915_guc_submission.c | 20 ++++++++++++++++----
  drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++
  drivers/gpu/drm/i915/intel_guc.c           |  2 ++
  5 files changed, 26 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -0
index e7e9e06..3df8a7d 100644
@@ -47,6 +47,7 @@
  #include <drm/i915_drm.h>
 
  #include "i915_drv.h"
+#include "i915_guc_submission.h"
  #include "i915_trace.h"
  #include "i915_vgpu.h"
  #include "intel_drv.h"
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct device *kdev)
 
Quoted 8 lines
Quoted 4 lines

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct device *kdev) · line 2619

Quoted 3 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
         intel_guc_resume(dev_priv);
 
+       i915_guc_clients_acquire_doorbells(&dev_priv->guc);

intel_guc_acquire_doorbells();

Prefixed "i915_guc_clients" since this modifies submission state (clients/doorbells). Should have kept dev_priv as parameter. what should be the correct option here: intel_guc*(guc) or i915_guc*(dev_priv)

GuC submission is not i915. It is not part of the user facing api. Operate on intel_guc as you were.
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw

Re: [PATCH 2/5] drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Quoted 59 lines · Chris Wilson

Quoting Sagar Arun Kamble (2017-11-07 06:05:01)

Quoted 58 lines · Chris Wilson
Quoted 3 lines

Quoting Sagar Arun Kamble (2017-11-05 13:39:40)

Quoted 2 lines

Before GT device suspend, i915 should release GuC client doorbells by stopping doorbell controller snooping and doorbell deallocation

through

Quoted 2 lines
Quoted 2 lines

GuC. They need to be acquired again on resume. This will also resolve the driver unload issue with GuC, where doorbell deallocation was

being

Quoted 2 lines
Quoted 2 lines

done post GEM suspend.
Release function is called from guc_suspend prior to sending sleep

action

Quoted 1 line
Quoted 1 line

during runtime and drm suspend. Acquiral is different in runtime and

drm

Quoted 1 line
Quoted 1 line

resume paths as on drm resume we are currently doing full reinit.

Release

Quoted 1 line
Quoted 1 line

should be done in synchronization with acquire hence GuC

suspend/resume

Quoted 1 line
Quoted 1 line

along with doorbell release/acquire should be under struct_mutex.

Upcoming

Quoted 13 lines
Quoted 9 lines

suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Quoted 4 lines
---
  drivers/gpu/drm/i915/i915_drv.c            |  3 +++
  drivers/gpu/drm/i915/i915_gem.c            |  5 +++--
  drivers/gpu/drm/i915/i915_guc_submission.c | 20  

++++++++++++++++----

Quoted 5 lines

Review

Quoted 5 lines
  drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++
  drivers/gpu/drm/i915/intel_guc.c           |  2 ++
  5 files changed, 26 insertions(+), 6 deletions(-)
(quoted diff) +0 -0

b/drivers/gpu/drm/i915/i915_drv.c

Quoted 6 lines
Quoted 6 lines

index e7e9e06..3df8a7d 100644

drivers/gpu/drm/i915/i915_drv.c +0 -0
@@ -47,6 +47,7 @@
  #include <drm/i915_drm.h>
 
Quoted 6 lines

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct device

Quoted 6 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
  #include "i915_drv.h"
+#include "i915_guc_submission.h"
  #include "i915_trace.h"
  #include "i915_vgpu.h"
  #include "intel_drv.h"
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct device  

*kdev)

Quoted 4 lines
Quoted 4 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
         intel_guc_resume(dev_priv);
 
+       i915_guc_clients_acquire_doorbells(&dev_priv->guc);
Quoted 8 lines
Quoted 5 lines
Quoted 1 line

intel_guc_acquire_doorbells();

Prefixed "i915_guc_clients" since this modifies submission state (clients/doorbells). Should have kept dev_priv as parameter. what should be the correct option here: intel_guc*(guc) or i915_guc*(dev_priv)

GuC submission is not i915. It is not part of the user facing api. Operate on intel_guc as you were.

So if GuC submission is not i915 than maybe to avoid confusion we should rename i915_guc_submission.c to intel_guc_submission.c ?

Michal
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#6 lore raw

Re: [PATCH 2/5] drm/i915/guc: Release all client doorbells on suspend and acquire on resume

Quoted 69 lines · Michal Wajdeczko
Quoted 67 lines · Chris Wilson

Quoting Sagar Arun Kamble (2017-11-07 06:05:01)

Quoted 65 lines · Chris Wilson
Quoted 2 lines

Quoting Sagar Arun Kamble (2017-11-05 13:39:40)

Quoted 1 line

Before GT device suspend, i915 should release GuC client

doorbells by

Quoted 1 line
Quoted 1 line

stopping doorbell controller snooping and doorbell deallocation

through

Quoted 1 line
Quoted 1 line

GuC. They need to be acquired again on resume. This will also

resolve

Quoted 1 line
Quoted 1 line

the driver unload issue with GuC, where doorbell deallocation was

being

Quoted 2 lines
Quoted 2 lines

done post GEM suspend.
Release function is called from guc_suspend prior to sending

sleep action

Quoted 1 line
Quoted 1 line

during runtime and drm suspend. Acquiral is different in runtime

and drm

Quoted 1 line
Quoted 1 line

resume paths as on drm resume we are currently doing full reinit.

Release

Quoted 1 line
Quoted 1 line

should be done in synchronization with acquire hence GuC

suspend/resume

Quoted 1 line
Quoted 1 line

along with doorbell release/acquire should be under struct_mutex.

Upcoming

Quoted 13 lines
Quoted 9 lines

suspend and resume restructuring for GuC will update these changes.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Quoted 4 lines

---
   drivers/gpu/drm/i915/i915_drv.c            |  3 +++ drivers/gpu/drm/i915/i915_gem.c            |  5 +++-- drivers/gpu/drm/i915/i915_guc_submission.c | 20

++++++++++++++++----

Quoted 5 lines

Review

Quoted 5 lines

   drivers/gpu/drm/i915/i915_guc_submission.h |  2 ++ drivers/gpu/drm/i915/intel_guc.c           |  2 ++

   5 files changed, 26 insertions(+), 6 deletions(-)
(quoted diff) +0 -0

b/drivers/gpu/drm/i915/i915_drv.c

Quoted 6 lines
Quoted 6 lines

index e7e9e06..3df8a7d 100644

drivers/gpu/drm/i915/i915_drv.c +0 -0
@@ -47,6 +47,7 @@
    #include <drm/i915_drm.h>
 
Quoted 6 lines

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct

Quoted 6 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
    #include "i915_drv.h"
+#include "i915_guc_submission.h"
    #include "i915_trace.h"
    #include "i915_vgpu.h"
    #include "intel_drv.h"
@@ -2615,6 +2616,8 @@ static int intel_runtime_resume(struct 

device *kdev)

Quoted 5 lines

drivers/gpu/drm/i915/i915_drv.c static int intel_runtime_resume(struct device *kdev) · line 2619

Quoted 4 lines
drivers/gpu/drm/i915/i915_drv.c +1 -0
           intel_guc_resume(dev_priv);
 
+ i915_guc_clients_acquire_doorbells(&dev_priv->guc);

intel_guc_acquire_doorbells();

Prefixed "i915_guc_clients" since this modifies submission state (clients/doorbells). Should have kept dev_priv as parameter. what should be the correct option here: intel_guc*(guc) or i915_guc*(dev_priv)

Quoted 5 lines
Quoted 2 lines

GuC submission is not i915. It is not part of the user facing api. Operate on intel_guc as you were.

So if GuC submission is not i915 than maybe to avoid confusion we should rename i915_guc_submission.c to intel_guc_submission.c ?

Yes. Will need to update functions that are i915_guc* as well. (spotted some in guc_log.c as well.)

Quoted 2 lines

Michal

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#7 lore raw

PATCH 3/5 drm/i915/guc: Deactivate all client doorbells before reset

In reply to Sagar Arun Kamble

Prior to i915 reset, we need to deactivate all client doorbells as they will need to be reacquired again post reset. We should not execute GuC doorbell deallocation flow as GuC might be in bad state (possibly reason for i915 reset).

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_gem.c            |  2 ++
 drivers/gpu/drm/i915/i915_guc_submission.c | 14 ++++++++++++++
 drivers/gpu/drm/i915/i915_guc_submission.h |  1 +
 3 files changed, 17 insertions(+)
drivers/gpu/drm/i915/i915_gem.c +2 -0
index cbce714..879e318 100644
@@ -2963,6 +2963,8 @@ int i915_gem_reset_prepare(struct drm_i915_private *dev_priv)
 
 	i915_gem_revoke_fences(dev_priv);
 
+	i915_guc_clients_reset_prepare(&dev_priv->guc);
+
 	return err;
 }
 
drivers/gpu/drm/i915/i915_guc_submission.c +14 -0
index b989edf..21f7fa7 100644
@@ -1113,6 +1113,20 @@ void i915_guc_clients_release_doorbells(struct intel_guc *guc)
 	destroy_doorbell(guc->execbuf_client, true);
 }
 
+void i915_guc_clients_reset_prepare(struct intel_guc *guc)
+{
+	if (!i915_modparams.enable_guc_submission)
+		return;
+
+	/*
+	 * We are only deactivating doorbell unit monitoring. GuC might
+	 * be in bad state and will be reset hence don't execute GuC
+	 * doorbell deallocation flow.
+	 */
+	destroy_doorbell(guc->preempt_client, false);
+	destroy_doorbell(guc->execbuf_client, false);
+}
+
 static void guc_policy_init(struct guc_policy *policy)
 {
 	policy->execution_quantum = POLICY_DEFAULT_EXECUTION_QUANTUM_US;
drivers/gpu/drm/i915/i915_guc_submission.h +1 -0
index 1d5cf22..91ad505 100644
@@ -74,6 +74,7 @@ struct i915_guc_client {
 
 void i915_guc_clients_acquire_doorbells(struct intel_guc *guc);
 void i915_guc_clients_release_doorbells(struct intel_guc *guc);
+void i915_guc_clients_reset_prepare(struct intel_guc *guc);
 int i915_guc_submission_init(struct drm_i915_private *dev_priv);
 int i915_guc_submission_enable(struct drm_i915_private *dev_priv);
 void i915_guc_submission_disable(struct drm_i915_private *dev_priv);
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#8 lore raw

PATCH 4/5 drm/i915/guc: Increase wait timeout for doorbell release status update

In reply to Sagar Arun Kamble

Current wait timeout of 10us is very tight as seen on SKL/KBL randomly for pm_rpm@basic-pci-d3-state. Increase it to 50us.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_guc_submission.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_guc_submission.c +2 -1
index 21f7fa7..3914415 100644
@@ -222,7 +222,8 @@ static int __destroy_doorbell(struct i915_guc_client *client, bool notify_guc)
 	/* Doorbell release flow requires that we wait for GEN8_DRB_VALID bit
 	 * to go to zero after updating db_status before we call the GuC to
 	 * release the doorbell */
-	if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID), 10))
+	if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID),
+			50))
 		WARN_ONCE(true, "Doorbell never became invalid after disable\n");
 
 	if (notify_guc)
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#9 lore raw

Re: [PATCH 4/5] drm/i915/guc: Increase wait timeout for doorbell release status update

Quoted 19 lines · Sagar Arun Kamble

Current wait timeout of 10us is very tight as seen on SKL/KBL randomly for pm_rpm@basic-pci-d3-state. Increase it to 50us.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>
---
 drivers/gpu/drm/i915/i915_guc_submission.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_guc_submission.c +0 -0
index 21f7fa7..3914415 100644

drivers/gpu/drm/i915/i915_guc_submission.c static int __destroy_doorbell(struct i915_guc_client · line 228

Quoted 8 lines
drivers/gpu/drm/i915/i915_guc_submission.c +2 -1
@@ -222,7 +222,8 @@ static int __destroy_doorbell(struct i915_guc_client  
 *client, bool notify_guc)
 	/* Doorbell release flow requires that we wait for GEN8_DRB_VALID bit
 	 * to go to zero after updating db_status before we call the GuC to
 	 * release the doorbell */
-	if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID),  
 10))
+	if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID),
+			50))

Can we use intel_wait_for_register() here ?

Quoted 2 lines
drivers/gpu/drm/i915/i915_guc_submission.c +0 -0
 		WARN_ONCE(true, "Doorbell never became invalid after disable\n");

if (notify_guc)

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#10 lore raw

Re: [PATCH 4/5] drm/i915/guc: Increase wait timeout for doorbell release status update

Quoted 16 lines · Michal Wajdeczko
Quoted 14 lines · Sagar Arun Kamble

Current wait timeout of 10us is very tight as seen on SKL/KBL randomly for pm_rpm@basic-pci-d3-state. Increase it to 50us.

Cc Chris Wilson <[email protected]>
Cc Michał Winiarski <[email protected]>
Cc Michel Thierry <[email protected]>
Cc Michal Wajdeczko <[email protected]>
Cc Arkadiusz Hiler <[email protected]>
Cc Joonas Lahtinen <[email protected]>

---
 drivers/gpu/drm/i915/i915_guc_submission.c | 3 ++-

 1 file changed, 2 insertions(+), 1 deletion(-)
Quoted 15 lines

drivers/gpu/drm/i915/i915_guc_submission.c static int __destroy_doorbell(struct · line 228

Quoted 13 lines
drivers/gpu/drm/i915/i915_guc_submission.c +2 -1
index 21f7fa7..3914415 100644
@@ -222,7 +222,8 @@ static int __destroy_doorbell(struct 
 i915_guc_client *client, bool notify_guc)
      /* Doorbell release flow requires that we wait for GEN8_DRB_VALID bit
       * to go to zero after updating db_status before we call the GuC to
       * release the doorbell */
-    if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & 
 GEN8_DRB_VALID), 10))
+    if (wait_for_us(!(I915_READ(GEN8_DRBREGL(db_id)) & GEN8_DRB_VALID),
+            50))

Can we use intel_wait_for_register() here ?

Yes. Since wait timeout is expected to be in the order of uS (50us for now) will be using __intel_wait_for_register_fw.

+        if (__intel_wait_for_register_fw(dev_priv, GEN8_DRBREGL(db_id), GEN8_DRB_VALID, 0,
+ 50, 1, NULL))

Is this fine (with slow timeout of 1ms)?

Quoted 2 lines
Quoted 2 lines
drivers/gpu/drm/i915/i915_guc_submission.c +0 -0
          WARN_ONCE(true, "Doorbell never became invalid after disable\n");

    if (notify_guc)

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#11 lore raw

PATCH 5/5 HAX enable GuC submission for CI

In reply to Sagar Arun Kamble

Also revert ("drm/i915/guc: Assert that we switch between known ggtt->invalidate functions")

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++------
 drivers/gpu/drm/i915/i915_params.h  | 4 ++--
 2 files changed, 4 insertions(+), 8 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +2 -6
index 0684d5d..a351ddf 100644
@@ -3562,17 +3562,13 @@ int i915_ggtt_enable_hw(struct drm_i915_private *dev_priv)
 
 void i915_ggtt_enable_guc(struct drm_i915_private *i915)
 {
-	GEM_BUG_ON(i915->ggtt.invalidate != gen6_ggtt_invalidate);
-
 	i915->ggtt.invalidate = guc_ggtt_invalidate;
 }
 
 void i915_ggtt_disable_guc(struct drm_i915_private *i915)
 {
-	/* We should only be called after i915_ggtt_enable_guc() */
-	GEM_BUG_ON(i915->ggtt.invalidate != guc_ggtt_invalidate);
-
-	i915->ggtt.invalidate = gen6_ggtt_invalidate;
+	if (i915->ggtt.invalidate == guc_ggtt_invalidate)
+		i915->ggtt.invalidate = gen6_ggtt_invalidate;
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_i915_private *dev_priv)
drivers/gpu/drm/i915/i915_params.h +2 -2
index c729226..c38cef0 100644
@@ -44,8 +44,8 @@
 	param(int, disable_power_well, -1) \
 	param(int, enable_ips, 1) \
 	param(int, invert_brightness, 0) \
-	param(int, enable_guc_loading, 0) \
-	param(int, enable_guc_submission, 0) \
+	param(int, enable_guc_loading, 1) \
+	param(int, enable_guc_submission, 1) \
 	param(int, guc_log_level, -1) \
 	param(char *, guc_firmware_path, NULL) \
 	param(char *, huc_firmware_path, NULL) \
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#12 lore raw

✗ Fi.CI.BAT: warning for series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

In reply to Sagar Arun Kamble
== Series Details ==

== Series Details ==

Series: series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task
URL : https://patchwork.freedesktop.org/series/33209/
State : warning

== Summary ==

Series 33209v1 series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task
https://patchwork.freedesktop.org/api/1.0/series/33209/revisions/1/mbox/

Test kms_pipe_crc_basic:

        Subgroup read-crc-pipe-c:
                dmesg-warn -> PASS       (fi-bsw-n3050)
Test drv_module_reload:
        Subgroup basic-reload:
                pass       -> DMESG-WARN (fi-cnl-y)
        Subgroup basic-no-display:
                pass       -> DMESG-WARN (fi-cnl-y)
        Subgroup basic-reload-inject:
                pass       -> DMESG-WARN (fi-cnl-y)
fi-bdw-5557u     total:289  pass:268  dwarn:0   dfail:0   fail:0   skip:21  time:444s
fi-bdw-gvtdvm    total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:454s
fi-blb-e6850     total:289  pass:223  dwarn:1   dfail:0   fail:0   skip:65  time:380s
fi-bsw-n3050     total:289  pass:243  dwarn:0   dfail:0   fail:0   skip:46  time:538s
fi-bwr-2160      total:289  pass:183  dwarn:0   dfail:0   fail:0   skip:106 time:276s
fi-bxt-dsi       total:289  pass:259  dwarn:0   dfail:0   fail:0   skip:30  time:510s
fi-bxt-j4205     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:512s
fi-byt-j1900     total:289  pass:253  dwarn:1   dfail:0   fail:0   skip:35  time:503s
fi-byt-n2820     total:289  pass:249  dwarn:1   dfail:0   fail:0   skip:39  time:486s
fi-cnl-y         total:289  pass:259  dwarn:3   dfail:0   fail:0   skip:27  time:626s
fi-elk-e7500     total:289  pass:229  dwarn:0   dfail:0   fail:0   skip:60  time:425s
fi-gdg-551       total:289  pass:178  dwarn:1   dfail:0   fail:1   skip:109 time:265s
fi-glk-1         total:289  pass:261  dwarn:0   dfail:0   fail:0   skip:28  time:588s
fi-hsw-4770      total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:435s
fi-hsw-4770r     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:430s
fi-ilk-650       total:289  pass:228  dwarn:0   dfail:0   fail:0   skip:61  time:428s
fi-ivb-3520m     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:497s
fi-ivb-3770      total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:466s
fi-kbl-7560u     total:289  pass:270  dwarn:0   dfail:0   fail:0   skip:19  time:574s
fi-kbl-7567u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:479s
fi-kbl-r         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:590s
fi-pnv-d510      total:289  pass:222  dwarn:1   dfail:0   fail:0   skip:66  time:577s
fi-skl-6260u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:450s
fi-skl-6600u     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:592s
fi-skl-6700hq    total:289  pass:263  dwarn:0   dfail:0   fail:0   skip:26  time:649s
fi-skl-6700k     total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:517s
fi-skl-6770hq    total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:502s
fi-skl-gvtdvm    total:289  pass:266  dwarn:0   dfail:0   fail:0   skip:23  time:464s
fi-snb-2520m     total:289  pass:250  dwarn:0   dfail:0   fail:0   skip:39  time:570s
fi-snb-2600      total:289  pass:249  dwarn:0   dfail:0   fail:0   skip:40  time:428s

8b0ae6b50a229dc661a02f4034252ee854cc9b83 drm-tip: 2017y-11m-03d-17h-15m-57s UTC integration manifest a91c31b1a24c HAX enable GuC submission for CI
32d332ce5a44 drm/i915/guc: Increase wait timeout for doorbell release status update be93da66d7df drm/i915/guc: Deactivate all client doorbells before reset
b6515b105ab5 drm/i915/guc: Release all client doorbells on suspend and acquire on resume 9e190e74bacd drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6966/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#13 lore raw

✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

In reply to Sagar Arun Kamble
== Series Details ==

== Series Details ==

Series: series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task
URL : https://patchwork.freedesktop.org/series/33209/
State : success

== Summary ==

Series 33209v1 series starting with [1/5] drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task
https://patchwork.freedesktop.org/api/1.0/series/33209/revisions/1/mbox/

Test kms_pipe_crc_basic:

        Subgroup read-crc-pipe-c:
                dmesg-warn -> PASS       (fi-bsw-n3050) fdo#103479

fdo#103479 https://bugs.freedesktop.org/show_bug.cgi?id=103479

fi-bdw-5557u     total:289  pass:268  dwarn:0   dfail:0   fail:0   skip:21  time:444s
fi-bdw-gvtdvm    total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:454s
fi-blb-e6850     total:289  pass:223  dwarn:1   dfail:0   fail:0   skip:65  time:380s
fi-bsw-n3050     total:289  pass:243  dwarn:0   dfail:0   fail:0   skip:46  time:538s
fi-bwr-2160      total:289  pass:183  dwarn:0   dfail:0   fail:0   skip:106 time:276s
fi-bxt-dsi       total:289  pass:259  dwarn:0   dfail:0   fail:0   skip:30  time:510s
fi-bxt-j4205     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:512s
fi-byt-j1900     total:289  pass:253  dwarn:1   dfail:0   fail:0   skip:35  time:503s
fi-byt-n2820     total:289  pass:249  dwarn:1   dfail:0   fail:0   skip:39  time:486s
fi-elk-e7500     total:289  pass:229  dwarn:0   dfail:0   fail:0   skip:60  time:425s
fi-gdg-551       total:289  pass:178  dwarn:1   dfail:0   fail:1   skip:109 time:265s
fi-glk-1         total:289  pass:261  dwarn:0   dfail:0   fail:0   skip:28  time:588s
fi-hsw-4770      total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:435s
fi-hsw-4770r     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:430s
fi-ilk-650       total:289  pass:228  dwarn:0   dfail:0   fail:0   skip:61  time:428s
fi-ivb-3520m     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:497s
fi-ivb-3770      total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:466s
fi-kbl-7560u     total:289  pass:270  dwarn:0   dfail:0   fail:0   skip:19  time:574s
fi-kbl-7567u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:479s
fi-kbl-r         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:590s
fi-pnv-d510      total:289  pass:222  dwarn:1   dfail:0   fail:0   skip:66  time:577s
fi-skl-6260u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:450s
fi-skl-6600u     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:592s
fi-skl-6700hq    total:289  pass:263  dwarn:0   dfail:0   fail:0   skip:26  time:649s
fi-skl-6700k     total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:517s
fi-skl-6770hq    total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:502s
fi-skl-gvtdvm    total:289  pass:266  dwarn:0   dfail:0   fail:0   skip:23  time:464s
fi-snb-2520m     total:289  pass:250  dwarn:0   dfail:0   fail:0   skip:39  time:570s
fi-snb-2600      total:289  pass:249  dwarn:0   dfail:0   fail:0   skip:40  time:428s

Blacklisted hosts:
fi-cnl-y total:289 pass:259 dwarn:3 dfail:0 fail:0 skip:27 time:626s

8b0ae6b50a229dc661a02f4034252ee854cc9b83 drm-tip: 2017y-11m-03d-17h-15m-57s UTC integration manifest a91c31b1a24c HAX enable GuC submission for CI
32d332ce5a44 drm/i915/guc: Increase wait timeout for doorbell release status update be93da66d7df drm/i915/guc: Deactivate all client doorbells before reset
b6515b105ab5 drm/i915/guc: Release all client doorbells on suspend and acquire on resume 9e190e74bacd drm/i915/guc: Separate GuC doorbell destroy function into doorbell unit and GuC task

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6966/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx