[PATCH 15/37] drm/amdkfd: Support enumerating non-GPU devices

[Felix Kuehling]

From: Harish Kasiviswanathan <Harish.Kasiviswanathan at amd.com>

Modify kfd_topology_enum_kfd_devices(..) function to support non-GPU
nodes. The function returned NULL when it encountered non-GPU (say CPU)
nodes. This caused kfd_ioctl_create_event and kfd_init_apertures to fail
for Intel + Tonga.

kfd_topology_enum_kfd_devices will now parse all the nodes and return
valid kfd_dev for nodes with GPU.

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan at amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling at amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c |  7 ++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_pasid.c       |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h        |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c    | 18 +++++++++++-------
 4 files changed, 19 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
index c59384b..7377513 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
@@ -300,9 +300,14 @@ int kfd_init_apertures(struct kfd_process *process)
 	struct kfd_process_device *pdd;
 
 	/*Iterating over all devices*/
-	while ((dev = kfd_topology_enum_kfd_devices(id)) != NULL &&
+	while (kfd_topology_enum_kfd_devices(id, &dev) == 0 &&
 		id < NUM_OF_SUPPORTED_GPUS) {
 
+		if (!dev) {
+			id++; /* Skip non GPU devices */
+			continue;
+		}
+
 		pdd = kfd_create_process_device_data(dev, process);
 		if (!pdd) {
 			pr_err("Failed to create process device data\n");
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
index d6a7961..15fff44 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
@@ -59,7 +59,7 @@ unsigned int kfd_pasid_alloc(void)
 		struct kfd_dev *dev = NULL;
 		unsigned int i = 0;
 
-		while ((dev = kfd_topology_enum_kfd_devices(i)) != NULL) {
+		while ((kfd_topology_enum_kfd_devices(i, &dev)) == 0) {
 			if (dev && dev->kfd2kgd) {
 				kfd2kgd = dev->kfd2kgd;
 				break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 0c96a6b..69a6206 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -670,7 +670,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu);
 int kfd_topology_remove_device(struct kfd_dev *gpu);
 struct kfd_dev *kfd_device_by_id(uint32_t gpu_id);
 struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev);
-struct kfd_dev *kfd_topology_enum_kfd_devices(uint8_t idx);
+int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev);
 
 /* Interrupts */
 int kfd_interrupt_init(struct kfd_dev *dev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index f64350b..b2d2b7e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -927,22 +927,26 @@ int kfd_topology_remove_device(struct kfd_dev *gpu)
 	return res;
 }
 
-/*
- * When idx is out of bounds, the function will return NULL
+/* kfd_topology_enum_kfd_devices - Enumerate through all devices in KFD
+ *	topology. If GPU device is found @idx, then valid kfd_dev pointer is
+ *	returned through @kdev
+ * Return -	0: On success (@kdev will be NULL for non GPU nodes)
+ *		-1: If end of list
  */
-struct kfd_dev *kfd_topology_enum_kfd_devices(uint8_t idx)
+int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev)
 {
 
 	struct kfd_topology_device *top_dev;
-	struct kfd_dev *device = NULL;
 	uint8_t device_idx = 0;
 
+	*kdev = NULL;
 	down_read(&topology_lock);
 
 	list_for_each_entry(top_dev, &topology_device_list, list) {
 		if (device_idx == idx) {
-			device = top_dev->gpu;
-			break;
+			*kdev = top_dev->gpu;
+			up_read(&topology_lock);
+			return 0;
 		}
 
 		device_idx++;
@@ -950,7 +954,7 @@ struct kfd_dev *kfd_topology_enum_kfd_devices(uint8_t idx)
 
 	up_read(&topology_lock);
 
-	return device;
+	return -1;
 
 }
 
-- 
2.7.4