amd-gfx.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
To: amd-gfx@lists.freedesktop.org
Cc: alexander.deucher@amd.com, felix.kuehling@amd.com,
	Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Subject: [Patch v3 3/4] drm/amdkfd: refactor runtime pm for baco
Date: Thu,  6 Feb 2020 19:09:10 -0500	[thread overview]
Message-ID: <20200207000911.19166-4-rajneesh.bhardwaj@amd.com> (raw)
In-Reply-To: <20200207000911.19166-1-rajneesh.bhardwaj@amd.com>

So far the kfd driver implemented same routines for runtime and system
wide suspend and resume (s2idle or mem). During system wide suspend the
kfd aquires an atomic lock that prevents any more user processes to
create queues and interact with kfd driver and amd gpu. This mechanism
created problem when amdgpu device is runtime suspended with BACO
enabled. Any application that relies on kfd driver fails to load because
the driver reports a locked kfd device since gpu is runtime suspended.

However, in an ideal case, when gpu is runtime  suspended the kfd driver
should be able to:

 - auto resume amdgpu driver whenever a client requests compute service
 - prevent runtime suspend for amdgpu  while kfd is in use

This change refactors the amdgpu and amdkfd drivers to support BACO and
runtime power management.

Reviewed-by: Oak Zeng <oak.zeng@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 12 +++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h |  8 ++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |  4 +--
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    | 29 +++++++++-------
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h      |  1 +
 drivers/gpu/drm/amd/amdkfd/kfd_process.c   | 40 ++++++++++++++++++++--
 6 files changed, 68 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 8609287620ea..314c4a2a0354 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -178,18 +178,18 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 		kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry);
 }
 
-void amdgpu_amdkfd_suspend(struct amdgpu_device *adev)
+void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm)
 {
 	if (adev->kfd.dev)
-		kgd2kfd_suspend(adev->kfd.dev);
+		kgd2kfd_suspend(adev->kfd.dev, run_pm);
 }
 
-int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
+int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm)
 {
 	int r = 0;
 
 	if (adev->kfd.dev)
-		r = kgd2kfd_resume(adev->kfd.dev);
+		r = kgd2kfd_resume(adev->kfd.dev, run_pm);
 
 	return r;
 }
@@ -713,11 +713,11 @@ void kgd2kfd_exit(void)
 {
 }
 
-void kgd2kfd_suspend(struct kfd_dev *kfd)
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 }
 
-int kgd2kfd_resume(struct kfd_dev *kfd)
+int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 47b0f2957d1f..9e8db702d878 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -122,8 +122,8 @@ struct amdkfd_process_info {
 int amdgpu_amdkfd_init(void);
 void amdgpu_amdkfd_fini(void);
 
-void amdgpu_amdkfd_suspend(struct amdgpu_device *adev);
-int amdgpu_amdkfd_resume(struct amdgpu_device *adev);
+void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm);
+int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm);
 void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 			const void *ih_ring_entry);
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev);
@@ -249,8 +249,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			 struct drm_device *ddev,
 			 const struct kgd2kfd_shared_resources *gpu_resources);
 void kgd2kfd_device_exit(struct kfd_dev *kfd);
-void kgd2kfd_suspend(struct kfd_dev *kfd);
-int kgd2kfd_resume(struct kfd_dev *kfd);
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm);
+int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm);
 int kgd2kfd_pre_reset(struct kfd_dev *kfd);
 int kgd2kfd_post_reset(struct kfd_dev *kfd);
 void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index f33c49ed4f94..18fa78806317 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3312,7 +3312,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 		}
 	}
 
-	amdgpu_amdkfd_suspend(adev);
+	amdgpu_amdkfd_suspend(adev, !fbcon);
 
 	amdgpu_ras_suspend(adev);
 
@@ -3396,7 +3396,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 			}
 		}
 	}
-	r = amdgpu_amdkfd_resume(adev);
+	r = amdgpu_amdkfd_resume(adev, !fbcon);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 798ad1c8f799..42ee9ea5c45a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -732,7 +732,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 void kgd2kfd_device_exit(struct kfd_dev *kfd)
 {
 	if (kfd->init_complete) {
-		kgd2kfd_suspend(kfd);
+		kgd2kfd_suspend(kfd, false);
 		device_queue_manager_uninit(kfd->dqm);
 		kfd_interrupt_exit(kfd);
 		kfd_topology_remove_device(kfd);
@@ -753,7 +753,7 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 
 	kfd->dqm->ops.pre_reset(kfd->dqm);
 
-	kgd2kfd_suspend(kfd);
+	kgd2kfd_suspend(kfd, false);
 
 	kfd_signal_reset_event(kfd);
 	return 0;
@@ -787,21 +787,23 @@ bool kfd_is_locked(void)
 	return  (atomic_read(&kfd_locked) > 0);
 }
 
-void kgd2kfd_suspend(struct kfd_dev *kfd)
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 	if (!kfd->init_complete)
 		return;
 
-	/* For first KFD device suspend all the KFD processes */
-	if (atomic_inc_return(&kfd_locked) == 1)
-		kfd_suspend_all_processes();
+	/* for runtime suspend, skip locking kfd */
+	if (!run_pm) {
+		/* For first KFD device suspend all the KFD processes */
+		if (atomic_inc_return(&kfd_locked) == 1)
+			kfd_suspend_all_processes();
+	}
 
 	kfd->dqm->ops.stop(kfd->dqm);
-
 	kfd_iommu_suspend(kfd);
 }
 
-int kgd2kfd_resume(struct kfd_dev *kfd)
+int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
 	int ret, count;
 
@@ -812,10 +814,13 @@ int kgd2kfd_resume(struct kfd_dev *kfd)
 	if (ret)
 		return ret;
 
-	count = atomic_dec_return(&kfd_locked);
-	WARN_ONCE(count < 0, "KFD suspend / resume ref. error");
-	if (count == 0)
-		ret = kfd_resume_all_processes();
+	/* for runtime resume, skip unlocking kfd */
+	if (!run_pm) {
+		count = atomic_dec_return(&kfd_locked);
+		WARN_ONCE(count < 0, "KFD suspend / resume ref. error");
+		if (count == 0)
+			ret = kfd_resume_all_processes();
+	}
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index f805f557c3c4..063096ec832d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -651,6 +651,7 @@ struct kfd_process_device {
 	 * function.
 	 */
 	bool already_dequeued;
+	bool runtime_inuse;
 
 	/* Is this process/pasid bound to this device? (amd_iommu_bind_pasid) */
 	enum kfd_pdd_bound bound;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 98dcbb96b2e2..6d6c25fe2677 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -31,6 +31,7 @@
 #include <linux/compat.h>
 #include <linux/mman.h>
 #include <linux/file.h>
+#include <linux/pm_runtime.h>
 #include "amdgpu_amdkfd.h"
 #include "amdgpu.h"
 
@@ -527,6 +528,16 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 		kfree(pdd->qpd.doorbell_bitmap);
 		idr_destroy(&pdd->alloc_idr);
 
+		/*
+		 * before destroying pdd, make sure to report availability
+		 * for auto suspend
+		 */
+		if (pdd->runtime_inuse) {
+			pm_runtime_mark_last_busy(pdd->dev->ddev->dev);
+			pm_runtime_put_autosuspend(pdd->dev->ddev->dev);
+			pdd->runtime_inuse = false;
+		}
+
 		kfree(pdd);
 	}
 }
@@ -844,6 +855,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 	pdd->process = p;
 	pdd->bound = PDD_UNBOUND;
 	pdd->already_dequeued = false;
+	pdd->runtime_inuse = false;
 	list_add(&pdd->per_device_list, &p->per_device_data);
 
 	/* Init idr used for memory handle translation */
@@ -933,15 +945,39 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev,
 		return ERR_PTR(-ENOMEM);
 	}
 
+	/*
+	 * signal runtime-pm system to auto resume and prevent
+	 * further runtime suspend once device pdd is created until
+	 * pdd is destroyed.
+	 */
+	if (!pdd->runtime_inuse) {
+		err = pm_runtime_get_sync(dev->ddev->dev);
+		if (err < 0)
+			return ERR_PTR(err);
+	}
+
 	err = kfd_iommu_bind_process_to_device(pdd);
 	if (err)
-		return ERR_PTR(err);
+		goto out;
 
 	err = kfd_process_device_init_vm(pdd, NULL);
 	if (err)
-		return ERR_PTR(err);
+		goto out;
+
+	if (!err)
+		/*
+		 * make sure that runtime_usage counter is incremented
+		 * just once per pdd
+		 */
+		pdd->runtime_inuse = true;
 
 	return pdd;
+
+out:
+	/* balance runpm reference count and exit with error */
+	pm_runtime_mark_last_busy(dev->ddev->dev);
+	pm_runtime_put_autosuspend(dev->ddev->dev);
+	return ERR_PTR(err);
 }
 
 struct kfd_process_device *kfd_get_first_process_device_data(
-- 
2.17.1

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2020-02-07  0:09 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-07  0:09 [Patch v3 0/4] Enable BACO with KFD Rajneesh Bhardwaj
2020-02-07  0:09 ` [Patch v3 1/4] drm/amdgpu: Fix missing error check in suspend Rajneesh Bhardwaj
2020-02-07  0:09 ` [Patch v3 2/4] drm/amdkfd: show warning when kfd is locked Rajneesh Bhardwaj
2020-02-07  0:09 ` Rajneesh Bhardwaj [this message]
2020-02-07 21:49   ` [Patch v3 3/4] drm/amdkfd: refactor runtime pm for baco Felix Kuehling
2020-02-07  0:09 ` [Patch v3 4/4] drm/amdgpu/runpm: enable runpm on baco capable VI+ asics Rajneesh Bhardwaj

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=20200207000911.19166-4-rajneesh.bhardwaj@amd.com \
    --to=rajneesh.bhardwaj@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=felix.kuehling@amd.com \
    /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;
as well as URLs for NNTP newsgroup(s).