Received-SPF: pass (google.com: domain of linux-kernel+bounces-154282-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org,
	stable@vger.kernel.org
Cc: Zhigang Luo <Zhigang.Luo@amd.com>,
	Felix Kuehling <felix.kuehling@amd.com>,
	Alex Deucher <alexander.deucher@amd.com>,
	Sasha Levin <sashal@kernel.org>,
	Felix.Kuehling@amd.com,
	christian.koenig@amd.com,
	Xinhui.Pan@amd.com,
	airlied@gmail.com,
	daniel@ffwll.ch,
	amd-gfx@lists.freedesktop.org,
	dri-devel@lists.freedesktop.org
Subject: [PATCH AUTOSEL 6.6 21/29] amd/amdkfd: sync all devices to wait all processes being evicted
Date: Mon, 22 Apr 2024 19:17:02 -0400
Message-ID: <20240422231730.1601976-21-sashal@kernel.org>
In-Reply-To: <20240422231730.1601976-1-sashal@kernel.org>
References: <20240422231730.1601976-1-sashal@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Zhigang Luo <Zhigang.Luo@amd.com>

[ Upstream commit d06af584be5a769d124b7302b32a033e9559761d ]

If there are more than one device doing reset in parallel, the first
device will call kfd_suspend_all_processes() to evict all processes
on all devices, this call takes time to finish. other device will
start reset and recover without waiting. if the process has not been
evicted before doing recover, it will be restored, then caused page
fault.

Signed-off-by: Zhigang Luo <Zhigang.Luo@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 93ce181eb3baa..913c70a0ef44f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -935,7 +935,6 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 	struct kfd_node *node;
 	int i;
-	int count;
 
 	if (!kfd->init_complete)
 		return;
@@ -943,12 +942,10 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 	/* for runtime suspend, skip locking kfd */
 	if (!run_pm) {
 		mutex_lock(&kfd_processes_mutex);
-		count = ++kfd_locked;
-		mutex_unlock(&kfd_processes_mutex);
-
 		/* For first KFD device suspend all the KFD processes */
-		if (count == 1)
+		if (++kfd_locked == 1)
 			kfd_suspend_all_processes();
+		mutex_unlock(&kfd_processes_mutex);
 	}
 
 	for (i = 0; i < kfd->num_nodes; i++) {
@@ -959,7 +956,7 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 
 int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
-	int ret, count, i;
+	int ret, i;
 
 	if (!kfd->init_complete)
 		return 0;
@@ -973,12 +970,10 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 	/* for runtime resume, skip unlocking kfd */
 	if (!run_pm) {
 		mutex_lock(&kfd_processes_mutex);
-		count = --kfd_locked;
-		mutex_unlock(&kfd_processes_mutex);
-
-		WARN_ONCE(count < 0, "KFD suspend / resume ref. error");
-		if (count == 0)
+		if (--kfd_locked == 0)
 			ret = kfd_resume_all_processes();
+		WARN_ONCE(kfd_locked < 0, "KFD suspend / resume ref. error");
+		mutex_unlock(&kfd_processes_mutex);
 	}
 
 	return ret;
-- 
2.43.0