Fixed reset event issues

Issues include:
	SWDEV-480250
	SWDEV-480255
	SWDEV-480248

Signed-off-by: gabrpham <Gabriel.Pham@amd.com>
Change-Id: Icf12211e4b136f26fce18f09a7bf8b7e9cd20691


[ROCm/rocm_smi_lib commit: 76ac0808fe]
This commit is contained in:
gabrpham
2024-11-20 14:12:06 -06:00
committed by Charis Poag
parent 125bdaf4f5
commit c20d2d30ec
4 changed files with 1424 additions and 157 deletions
File diff suppressed because it is too large Load Diff
@@ -318,9 +318,15 @@ typedef enum {
RSMI_EVT_NOTIF_THERMAL_THROTTLE = KFD_SMI_EVENT_THERMAL_THROTTLE,
RSMI_EVT_NOTIF_GPU_PRE_RESET = KFD_SMI_EVENT_GPU_PRE_RESET,
RSMI_EVT_NOTIF_GPU_POST_RESET = KFD_SMI_EVENT_GPU_POST_RESET,
RSMI_EVT_NOTIF_RING_HANG = KFD_SMI_EVENT_RING_HANG,
RSMI_EVT_NOTIF_LAST = RSMI_EVT_NOTIF_RING_HANG
RSMI_EVT_NOTIF_EVENT_MIGRATE_START = KFD_SMI_EVENT_MIGRATE_START,
RSMI_EVT_NOTIF_EVENT_MIGRATE_END = KFD_SMI_EVENT_MIGRATE_END,
RSMI_EVT_NOTIF_EVENT_PAGE_FAULT_START = KFD_SMI_EVENT_PAGE_FAULT_START,
RSMI_EVT_NOTIF_EVENT_PAGE_FAULT_END = KFD_SMI_EVENT_PAGE_FAULT_END,
RSMI_EVT_NOTIF_EVENT_QUEUE_EVICTION = KFD_SMI_EVENT_QUEUE_EVICTION,
RSMI_EVT_NOTIF_EVENT_QUEUE_RESTORE = KFD_SMI_EVENT_QUEUE_RESTORE,
RSMI_EVT_NOTIF_EVENT_UNMAP_FROM_GPU = KFD_SMI_EVENT_UNMAP_FROM_GPU,
RSMI_EVT_NOTIF_EVENT_ALL_PROCESS = KFD_SMI_EVENT_ALL_PROCESS,
RSMI_EVT_NOTIF_LAST = KFD_SMI_EVENT_ALL_PROCESS
} rsmi_evt_notification_type_t;
/**
@@ -329,7 +335,8 @@ typedef enum {
#define RSMI_EVENT_MASK_FROM_INDEX(i) (1ULL << ((i) - 1))
//! Maximum number of characters an event notification message will be
#define MAX_EVENT_NOTIFICATION_MSG_SIZE 64
// matches kfd message max size
#define MAX_EVENT_NOTIFICATION_MSG_SIZE 96
/**
* Event notification data returned from event notification API