Sysman API¶
oneAPI Level Zero Specification - Version 1.3.7
Common¶
Enumerations
Structures
Common Enums¶
zes_structure_type_t¶
-
enum
zes_structure_type_t
¶ Defines structure types.
Values:
-
ZES_STRUCTURE_TYPE_DEVICE_PROPERTIES
= 0x1¶
-
ZES_STRUCTURE_TYPE_PCI_PROPERTIES
= 0x2¶
-
ZES_STRUCTURE_TYPE_PCI_BAR_PROPERTIES
= 0x3¶
-
ZES_STRUCTURE_TYPE_DIAG_PROPERTIES
= 0x4¶
-
ZES_STRUCTURE_TYPE_ENGINE_PROPERTIES
= 0x5¶
-
ZES_STRUCTURE_TYPE_FABRIC_PORT_PROPERTIES
= 0x6¶
-
ZES_STRUCTURE_TYPE_FAN_PROPERTIES
= 0x7¶
-
ZES_STRUCTURE_TYPE_FIRMWARE_PROPERTIES
= 0x8¶
-
ZES_STRUCTURE_TYPE_FREQ_PROPERTIES
= 0x9¶
-
ZES_STRUCTURE_TYPE_LED_PROPERTIES
= 0xa¶
-
ZES_STRUCTURE_TYPE_MEM_PROPERTIES
= 0xb¶
-
ZES_STRUCTURE_TYPE_PERF_PROPERTIES
= 0xc¶
-
ZES_STRUCTURE_TYPE_POWER_PROPERTIES
= 0xd¶
-
ZES_STRUCTURE_TYPE_PSU_PROPERTIES
= 0xe¶
-
ZES_STRUCTURE_TYPE_RAS_PROPERTIES
= 0xf¶
-
ZES_STRUCTURE_TYPE_SCHED_PROPERTIES
= 0x10¶
-
ZES_STRUCTURE_TYPE_SCHED_TIMEOUT_PROPERTIES
= 0x11¶
-
ZES_STRUCTURE_TYPE_SCHED_TIMESLICE_PROPERTIES
= 0x12¶
-
ZES_STRUCTURE_TYPE_STANDBY_PROPERTIES
= 0x13¶
-
ZES_STRUCTURE_TYPE_TEMP_PROPERTIES
= 0x14¶
-
ZES_STRUCTURE_TYPE_DEVICE_STATE
= 0x15¶
-
ZES_STRUCTURE_TYPE_PROCESS_STATE
= 0x16¶
-
ZES_STRUCTURE_TYPE_PCI_STATE
= 0x17¶
-
ZES_STRUCTURE_TYPE_FABRIC_PORT_CONFIG
= 0x18¶
-
ZES_STRUCTURE_TYPE_FABRIC_PORT_STATE
= 0x19¶
-
ZES_STRUCTURE_TYPE_FAN_CONFIG
= 0x1a¶
-
ZES_STRUCTURE_TYPE_FREQ_STATE
= 0x1b¶
-
ZES_STRUCTURE_TYPE_OC_CAPABILITIES
= 0x1c¶
-
ZES_STRUCTURE_TYPE_LED_STATE
= 0x1d¶
-
ZES_STRUCTURE_TYPE_MEM_STATE
= 0x1e¶
-
ZES_STRUCTURE_TYPE_PSU_STATE
= 0x1f¶
-
ZES_STRUCTURE_TYPE_BASE_STATE
= 0x20¶
-
ZES_STRUCTURE_TYPE_RAS_CONFIG
= 0x21¶
-
ZES_STRUCTURE_TYPE_RAS_STATE
= 0x22¶
-
ZES_STRUCTURE_TYPE_TEMP_CONFIG
= 0x23¶
-
ZES_STRUCTURE_TYPE_PCI_BAR_PROPERTIES_1_2
= 0x24¶
-
ZES_STRUCTURE_TYPE_FORCE_UINT32
= 0x7fffffff¶
-
Common Structures¶
zes_base_properties_t¶
-
struct
zes_base_properties_t
¶ Base for all properties types.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_structure_type_t
zes_base_desc_t¶
-
struct
zes_base_desc_t
¶ Base for all descriptor types.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_structure_type_t
zes_base_state_t¶
-
struct
zes_base_state_t
¶ Base for all state types.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_structure_type_t
zes_base_config_t¶
-
struct
zes_base_config_t
¶ Base for all config types.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_structure_type_t
zes_base_capability_t¶
-
struct
zes_base_capability_t
¶ Base for all capability types.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_structure_type_t
Device¶
Functions
Enumerations
Structures
Device Functions¶
zesDeviceGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceGetProperties
(zes_device_handle_t hDevice, zes_device_properties_t *pProperties)¶ Get properties about the device.
- Parameters
hDevice
: [in] Sysman handle of the device.pProperties
: [in,out] Structure that will contain information about the device.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesDeviceGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceGetState
(zes_device_handle_t hDevice, zes_device_state_t *pState)¶ Get information about the state of the device - if a reset is required, reasons for the reset and if the device has been repaired.
- Parameters
hDevice
: [in] Sysman handle of the device.pState
: [in,out] Structure that will contain information about the device.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesDeviceReset¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceReset
(zes_device_handle_t hDevice, ze_bool_t force)¶ Reset device.
- Parameters
hDevice
: [in] Sysman handle for the deviceforce
: [in] If set to true, all applications that are currently using the device will be forcibly killed.
Performs a PCI bus reset of the device. This will result in all current device state being lost.
All applications using the device should be stopped before calling this function.
If the force argument is specified, all applications using the device will be forcibly killed.
The function will block until the device has restarted or a timeout occurred waiting for the reset to complete.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to perform this operation.
ZE_RESULT_ERROR_HANDLE_OBJECT_IN_USE - “Reset cannot be performed because applications are using this device.”
ZE_RESULT_ERROR_UNKNOWN - “There were problems unloading the device driver, performing a bus reset or reloading the device driver.”
zesDeviceProcessesGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceProcessesGetState
(zes_device_handle_t hDevice, uint32_t *pCount, zes_process_state_t *pProcesses)¶ Get information about host processes using the device.
- Parameters
hDevice
: [in] Sysman handle for the devicepCount
: [in,out] pointer to the number of processes. if count is zero, then the driver shall update the value with the total number of processes currently attached to the device. if count is greater than the number of processes currently attached to the device, then the driver shall update the value with the correct number of processes.pProcesses
: [in,out][optional][range(0, *pCount)] array of process information. if count is less than the number of processes currently attached to the device, then the driver shall only retrieve information about that number of processes. In this case, the return code will ZE_RESULT_ERROR_INVALID_SIZE.
The number of processes connected to the device is dynamic. This means that between a call to determine the value of pCount and the subsequent call, the number of processes may have increased or decreased. It is recommended that a large array be passed in so as to avoid receiving the error ZE_RESULT_ERROR_INVALID_SIZE. Also, always check the returned value in pCount since it may be less than the earlier call to get the required array size.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
The provided value of pCount is not big enough to store information about all the processes currently attached to the device.
zesDevicePciGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDevicePciGetProperties
(zes_device_handle_t hDevice, zes_pci_properties_t *pProperties)¶ Get PCI properties - address, max speed.
- Parameters
hDevice
: [in] Sysman handle of the device.pProperties
: [in,out] Will contain the PCI properties.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesDevicePciGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDevicePciGetState
(zes_device_handle_t hDevice, zes_pci_state_t *pState)¶ Get current PCI state - current speed.
- Parameters
hDevice
: [in] Sysman handle of the device.pState
: [in,out] Will contain the PCI properties.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesDevicePciGetBars¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDevicePciGetBars
(zes_device_handle_t hDevice, uint32_t *pCount, zes_pci_bar_properties_t *pProperties)¶ Get information about each configured bar.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of PCI bars. if count is zero, then the driver shall update the value with the total number of PCI bars that are setup. if count is greater than the number of PCI bars that are setup, then the driver shall update the value with the correct number of PCI bars.pProperties
: [in,out][optional][range(0, *pCount)] array of information about setup PCI bars. if count is less than the number of PCI bars that are setup, then the driver shall only retrieve information about that number of PCI bars.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesDevicePciGetStats¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDevicePciGetStats
(zes_device_handle_t hDevice, zes_pci_stats_t *pStats)¶ Get PCI stats - bandwidth, number of packets, number of replays.
- Parameters
hDevice
: [in] Sysman handle of the device.pStats
: [in,out] Will contain a snapshot of the latest stats.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pStats
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to query this telemetry.
Device Enums¶
zes_engine_type_flags_t¶
-
enum
zes_engine_type_flag_t
¶ Values:
-
ZES_ENGINE_TYPE_FLAG_OTHER
= ZE_BIT(0)¶ Undefined types of accelerators.
-
ZES_ENGINE_TYPE_FLAG_COMPUTE
= ZE_BIT(1)¶ Engines that process compute kernels only (no 3D content).
-
ZES_ENGINE_TYPE_FLAG_3D
= ZE_BIT(2)¶ Engines that process 3D content only (no compute kernels).
-
ZES_ENGINE_TYPE_FLAG_MEDIA
= ZE_BIT(3)¶ Engines that process media workloads.
-
ZES_ENGINE_TYPE_FLAG_DMA
= ZE_BIT(4)¶ Engines that copy blocks of data.
-
ZES_ENGINE_TYPE_FLAG_RENDER
= ZE_BIT(5)¶ Engines that can process both 3D content and compute kernels.
-
ZES_ENGINE_TYPE_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
zes_repair_status_t¶
-
enum
zes_repair_status_t
¶ Device repair status.
Values:
-
ZES_REPAIR_STATUS_UNSUPPORTED
= 0¶ The device does not support in-field repairs.
-
ZES_REPAIR_STATUS_NOT_PERFORMED
= 1¶ The device has never been repaired.
-
ZES_REPAIR_STATUS_PERFORMED
= 2¶ The device has been repaired.
-
ZES_REPAIR_STATUS_FORCE_UINT32
= 0x7fffffff¶
-
zes_reset_reason_flags_t¶
-
enum
zes_reset_reason_flag_t
¶ Values:
-
ZES_RESET_REASON_FLAG_WEDGED
= ZE_BIT(0)¶ The device needs to be reset because one or more parts of the hardware is wedged
-
ZES_RESET_REASON_FLAG_REPAIR
= ZE_BIT(1)¶ The device needs to be reset in order to complete in-field repairs.
-
ZES_RESET_REASON_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
zes_pci_link_status_t¶
-
enum
zes_pci_link_status_t
¶ PCI link status.
Values:
-
ZES_PCI_LINK_STATUS_UNKNOWN
= 0¶ The link status could not be determined.
-
ZES_PCI_LINK_STATUS_GOOD
= 1¶ The link is up and operating as expected.
-
ZES_PCI_LINK_STATUS_QUALITY_ISSUES
= 2¶ The link is up but has quality and/or bandwidth degradation.
-
ZES_PCI_LINK_STATUS_STABILITY_ISSUES
= 3¶ The link has stability issues and preventing workloads making forward progress
-
ZES_PCI_LINK_STATUS_FORCE_UINT32
= 0x7fffffff¶
-
zes_pci_link_qual_issue_flags_t¶
-
enum
zes_pci_link_qual_issue_flag_t
¶ Values:
-
ZES_PCI_LINK_QUAL_ISSUE_FLAG_REPLAYS
= ZE_BIT(0)¶ A significant number of replays are occurring.
-
ZES_PCI_LINK_QUAL_ISSUE_FLAG_SPEED
= ZE_BIT(1)¶ There is a degradation in the maximum bandwidth of the link.
-
ZES_PCI_LINK_QUAL_ISSUE_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
zes_pci_link_stab_issue_flags_t¶
Device Structures¶
zes_device_state_t¶
-
struct
zes_device_state_t
¶ Device state.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_reset_reason_flags_t
reset
¶ [out] Indicates if the device needs to be reset and for what reasons. returns 0 (none) or combination of zes_reset_reason_flag_t
-
zes_repair_status_t
repaired
¶ [out] Indicates if the device has been repaired
-
zes_structure_type_t
zes_device_properties_t¶
-
struct
zes_device_properties_t
¶ Device properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_device_properties_t
core
¶ [out] Core device properties
-
uint32_t
numSubdevices
¶ [out] Number of sub-devices. A value of 0 indicates that this device doesn’t have sub-devices.
-
char
serialNumber
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Manufacturing serial number (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
char
boardNumber
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Manufacturing board number (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
char
brandName
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Brand name of the device (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
char
modelName
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Model name of the device (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
char
vendorName
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Vendor name of the device (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
char
driverVersion
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Installed driver version (NULL terminated string value). Will be set to the string “unkown” if this cannot be determined for the device.
-
zes_structure_type_t
zes_process_state_t¶
-
struct
zes_process_state_t
¶ Contains information about a process that has an open connection with this device.
The application can use the process ID to query the OS for the owner and the path to the executable.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
uint32_t
processId
¶ [out] Host OS process ID.
-
uint64_t
memSize
¶ [out] Device memory size in bytes allocated by this process (may not necessarily be resident on the device at the time of reading).
[out] The size of shared device memory mapped into this process (may not necessarily be resident on the device at the time of reading).
-
zes_engine_type_flags_t
engines
¶ [out] Bitfield of accelerator engine types being used by this process.
zes_pci_speed_t¶
-
struct
zes_pci_speed_t
¶ PCI speed.
Public Members
-
int32_t
gen
¶ [out] The link generation. A value of -1 means that this property is unknown.
-
int32_t
width
¶ [out] The number of lanes. A value of -1 means that this property is unknown.
-
int64_t
maxBandwidth
¶ [out] The maximum bandwidth in bytes/sec (sum of all lanes). A value of -1 means that this property is unknown.
-
int32_t
zes_pci_properties_t¶
-
struct
zes_pci_properties_t
¶ Static PCI properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_pci_address_t
address
¶ [out] The BDF address
-
zes_pci_speed_t
maxSpeed
¶ [out] Fastest port configuration supported by the device (sum of all lanes)
-
ze_bool_t
haveBandwidthCounters
¶ [out] Indicates if zes_pci_stats_t.rxCounter and zes_pci_stats_t.txCounter will have valid values
-
ze_bool_t
havePacketCounters
¶ [out] Indicates if zes_pci_stats_t.packetCounter will have valid values
-
ze_bool_t
haveReplayCounters
¶ [out] Indicates if zes_pci_stats_t.replayCounter will have valid values
-
zes_structure_type_t
zes_pci_state_t¶
-
struct
zes_pci_state_t
¶ Dynamic PCI state.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_pci_link_status_t
status
¶ [out] The current status of the port
-
zes_pci_link_qual_issue_flags_t
qualityIssues
¶ [out] If status is ZES_PCI_LINK_STATUS_QUALITY_ISSUES, then this gives a combination of zes_pci_link_qual_issue_flag_t for quality issues that have been detected; otherwise, 0 indicates there are no quality issues with the link at this time.”
-
zes_pci_link_stab_issue_flags_t
stabilityIssues
¶ [out] If status is ZES_PCI_LINK_STATUS_STABILITY_ISSUES, then this gives a combination of zes_pci_link_stab_issue_flag_t for reasons for the connection instability; otherwise, 0 indicates there are no connection stability issues at this time.”
-
zes_pci_speed_t
speed
¶ [out] The current port configure speed
-
zes_structure_type_t
zes_pci_bar_properties_t¶
-
struct
zes_pci_bar_properties_t
¶ Properties of a pci bar.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_pci_bar_type_t
type
¶ [out] The type of bar
-
uint32_t
index
¶ [out] The index of the bar
-
uint64_t
base
¶ [out] Base address of the bar.
-
uint64_t
size
¶ [out] Size of the bar.
-
zes_structure_type_t
zes_pci_bar_properties_1_2_t¶
-
struct
zes_pci_bar_properties_1_2_t
¶ Properties of a pci bar, including the resizable bar.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_pci_bar_type_t
type
¶ [out] The type of bar
-
uint32_t
index
¶ [out] The index of the bar
-
uint64_t
base
¶ [out] Base address of the bar.
-
uint64_t
size
¶ [out] Size of the bar.
-
ze_bool_t
resizableBarSupported
¶ [out] Support for Resizable Bar on this device.
-
ze_bool_t
resizableBarEnabled
¶ [out] Resizable Bar enabled on this device
-
zes_structure_type_t
zes_pci_stats_t¶
-
struct
zes_pci_stats_t
¶ PCI stats counters.
Percent replays is calculated by taking two snapshots (s1, s2) and using the equation: replay = 10^6 * (s2.replayCounter - s1.replayCounter) / (s2.maxBandwidth * (s2.timestamp - s1.timestamp))
Percent throughput is calculated by taking two snapshots (s1, s2) and using the equation: bw = 10^6 * ((s2.rxCounter - s1.rxCounter) + (s2.txCounter - s1.txCounter)) / (s2.maxBandwidth * (s2.timestamp - s1.timestamp))
Public Members
-
uint64_t
timestamp
¶ [out] Monotonic timestamp counter in microseconds when the measurement was made. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
-
uint64_t
replayCounter
¶ [out] Monotonic counter for the number of replay packets (sum of all lanes). Will always be 0 if zes_pci_properties_t.haveReplayCounters is FALSE.
-
uint64_t
packetCounter
¶ [out] Monotonic counter for the number of packets (sum of all lanes). Will always be 0 if zes_pci_properties_t.havePacketCounters is FALSE.
-
uint64_t
rxCounter
¶ [out] Monotonic counter for the number of bytes received (sum of all lanes). Will always be 0 if zes_pci_properties_t.haveBandwidthCounters is FALSE.
-
uint64_t
txCounter
¶ [out] Monotonic counter for the number of bytes transmitted (including replays) (sum of all lanes). Will always be 0 if zes_pci_properties_t.haveBandwidthCounters is FALSE.
-
zes_pci_speed_t
speed
¶ [out] The current speed of the link (sum of all lanes)
Diagnostics¶
Functions
Enumerations
Structures
Diagnostics Functions¶
zesDeviceEnumDiagnosticTestSuites¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumDiagnosticTestSuites
(zes_device_handle_t hDevice, uint32_t *pCount, zes_diag_handle_t *phDiagnostics)¶ Get handle of diagnostics test suites.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phDiagnostics
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesDiagnosticsGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDiagnosticsGetProperties
(zes_diag_handle_t hDiagnostics, zes_diag_properties_t *pProperties)¶ Get properties of a diagnostics test suite.
- Parameters
hDiagnostics
: [in] Handle for the component.pProperties
: [in,out] Structure describing the properties of a diagnostics test suite
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDiagnostics
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesDiagnosticsGetTests¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDiagnosticsGetTests
(zes_diag_handle_t hDiagnostics, uint32_t *pCount, zes_diag_test_t *pTests)¶ Get individual tests that can be run separately. Not all test suites permit running individual tests - check zes_diag_properties_t.haveTests.
- Parameters
hDiagnostics
: [in] Handle for the component.pCount
: [in,out] pointer to the number of tests. if count is zero, then the driver shall update the value with the total number of tests that are available. if count is greater than the number of tests that are available, then the driver shall update the value with the correct number of tests.pTests
: [in,out][optional][range(0, *pCount)] array of information about individual tests sorted by increasing value of zes_diag_test_t.index. if count is less than the number of tests that are available, then the driver shall only retrieve that number of tests.
The list of available tests is returned in order of increasing test index zes_diag_test_t.index.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDiagnostics
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesDiagnosticsRunTests¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDiagnosticsRunTests
(zes_diag_handle_t hDiagnostics, uint32_t startIndex, uint32_t endIndex, zes_diag_result_t *pResult)¶ Run a diagnostics test suite, either all tests or a subset of tests.
- Parameters
hDiagnostics
: [in] Handle for the component.startIndex
: [in] The index of the first test to run. Set to ZES_DIAG_FIRST_TEST_INDEX to start from the beginning.endIndex
: [in] The index of the last test to run. Set to ZES_DIAG_LAST_TEST_INDEX to complete all tests after the start test.pResult
: [in,out] The result of the diagnostics
WARNING: Running diagnostics may destroy current device state information. Gracefully close any running workloads before initiating.
To run all tests in a test suite, set start = ZES_DIAG_FIRST_TEST_INDEX and end = ZES_DIAG_LAST_TEST_INDEX.
If the test suite permits running individual tests, zes_diag_properties_t.haveTests will be true. In this case, the function zesDiagnosticsGetTests() can be called to get the list of tests and corresponding indices that can be supplied to the arguments start and end in this function.
This function will block until the diagnostics have completed and force reset based on result
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDiagnostics
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pResult
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to perform diagnostics.
Diagnostics Enums¶
zes_diag_result_t¶
-
enum
zes_diag_result_t
¶ Diagnostic results.
Values:
-
ZES_DIAG_RESULT_NO_ERRORS
= 0¶ Diagnostic completed without finding errors to repair.
-
ZES_DIAG_RESULT_ABORT
= 1¶ Diagnostic had problems running tests.
-
ZES_DIAG_RESULT_FAIL_CANT_REPAIR
= 2¶ Diagnostic had problems setting up repairs.
-
ZES_DIAG_RESULT_REBOOT_FOR_REPAIR
= 3¶ Diagnostics found errors, setup for repair and reboot is required to complete the process
-
ZES_DIAG_RESULT_FORCE_UINT32
= 0x7fffffff¶
-
Diagnostics Structures¶
zes_diag_properties_t¶
-
struct
zes_diag_properties_t
¶ Diagnostics test suite properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
char
name
[ZES_STRING_PROPERTY_SIZE
]¶ [out] Name of the diagnostics test suite
-
ze_bool_t
haveTests
¶ [out] Indicates if this test suite has individual tests which can be run separately (use the function zesDiagnosticsGetTests() to get the list of these tests)
-
zes_structure_type_t
Engine¶
Functions
Enumerations
Structures
Engine Functions¶
zesDeviceEnumEngineGroups¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumEngineGroups
(zes_device_handle_t hDevice, uint32_t *pCount, zes_engine_handle_t *phEngine)¶ Get handle of engine groups.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phEngine
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesEngineGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesEngineGetProperties
(zes_engine_handle_t hEngine, zes_engine_properties_t *pProperties)¶ Get engine group properties.
- Parameters
hEngine
: [in] Handle for the component.pProperties
: [in,out] The properties for the specified engine group.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hEngine
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesEngineGetActivity¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesEngineGetActivity
(zes_engine_handle_t hEngine, zes_engine_stats_t *pStats)¶ Get the activity stats for an engine group.
- Parameters
hEngine
: [in] Handle for the component.pStats
: [in,out] Will contain a snapshot of the engine group activity counters.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hEngine
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pStats
Engine Enums¶
zes_engine_group_t¶
-
enum
zes_engine_group_t
¶ Accelerator engine groups.
Values:
-
ZES_ENGINE_GROUP_ALL
= 0¶ Access information about all engines combined.
-
ZES_ENGINE_GROUP_COMPUTE_ALL
= 1¶ Access information about all compute engines combined. Compute engines can only process compute kernels (no 3D content).
-
ZES_ENGINE_GROUP_MEDIA_ALL
= 2¶ Access information about all media engines combined.
-
ZES_ENGINE_GROUP_COPY_ALL
= 3¶ Access information about all copy (blitter) engines combined.
-
ZES_ENGINE_GROUP_COMPUTE_SINGLE
= 4¶ Access information about a single compute engine - this is an engine that can process compute kernels. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_3D_RENDER_COMPUTE_ALL for that.
-
ZES_ENGINE_GROUP_RENDER_SINGLE
= 5¶ Access information about a single render engine - this is an engine that can process both 3D content and compute kernels. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_3D_RENDER_COMPUTE_ALL for that.
-
ZES_ENGINE_GROUP_MEDIA_DECODE_SINGLE
= 6¶ Access information about a single media decode engine. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_MEDIA_ALL for that.
-
ZES_ENGINE_GROUP_MEDIA_ENCODE_SINGLE
= 7¶ Access information about a single media encode engine. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_MEDIA_ALL for that.
-
ZES_ENGINE_GROUP_COPY_SINGLE
= 8¶ Access information about a single media encode engine. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_COPY_ALL for that.
-
ZES_ENGINE_GROUP_MEDIA_ENHANCEMENT_SINGLE
= 9¶ Access information about a single media enhancement engine. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_MEDIA_ALL for that.
-
ZES_ENGINE_GROUP_3D_SINGLE
= 10¶ Access information about a single 3D engine - this is an engine that can process 3D content only. Note that single engines may share the same underlying accelerator resources as other engines so activity of such an engine may not be indicative of the underlying resource utilization - use ZES_ENGINE_GROUP_3D_RENDER_COMPUTE_ALL for that.
-
ZES_ENGINE_GROUP_3D_RENDER_COMPUTE_ALL
= 11¶ Access information about all 3D/render/compute engines combined.
-
ZES_ENGINE_GROUP_RENDER_ALL
= 12¶ Access information about all render engines combined. Render engines are those than process both 3D content and compute kernels.
-
ZES_ENGINE_GROUP_3D_ALL
= 13¶ Access information about all 3D engines combined. 3D engines can process 3D content only (no compute kernels).
-
ZES_ENGINE_GROUP_FORCE_UINT32
= 0x7fffffff¶
-
Engine Structures¶
zes_engine_properties_t¶
-
struct
zes_engine_properties_t
¶ Engine group properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_engine_group_t
type
¶ [out] The engine group
-
ze_bool_t
onSubdevice
¶ [out] True if this resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_structure_type_t
zes_engine_stats_t¶
-
struct
zes_engine_stats_t
¶ Engine activity counters.
Percent utilization is calculated by taking two snapshots (s1, s2) and using the equation: util = (s2.activeTime - s1.activeTime) / (s2.timestamp - s1.timestamp)
Public Members
-
uint64_t
activeTime
¶ [out] Monotonic counter for time in microseconds that this resource is actively running workloads.
-
uint64_t
timestamp
¶ [out] Monotonic timestamp counter in microseconds when activeTime counter was sampled. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
Events¶
Functions
Enumerations
Events Functions¶
zesDeviceEventRegister¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEventRegister
(zes_device_handle_t hDevice, zes_event_type_flags_t events)¶ Specify the list of events to listen to for a given device.
- Parameters
hDevice
: [in] The device handle.events
: [in] List of events to listen to.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_ENUMERATION
0x7fff < events
zesDriverEventListen¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDriverEventListen
(ze_driver_handle_t hDriver, uint32_t timeout, uint32_t count, zes_device_handle_t *phDevices, uint32_t *pNumDeviceEvents, zes_event_type_flags_t *pEvents)¶ Wait for events to be received from a one or more devices.
- Parameters
hDriver
: [in] handle of the driver instancetimeout
: [in] if non-zero, then indicates the maximum time (in milliseconds) to yield before returning ZE_RESULT_SUCCESS or ZE_RESULT_NOT_READY; if zero, then will check status and return immediately; if UINT32_MAX, then function will not return until events arrive.count
: [in] Number of device handles in phDevices.phDevices
: [in][range(0, count)] Device handles to listen to for events. Only devices from the provided driver handle can be specified in this list.pNumDeviceEvents
: [in,out] Will contain the actual number of devices in phDevices that generated events. If non-zero, check pEvents to determine the devices and events that were received.pEvents
: [in,out] An array that will continue the list of events for each device listened in phDevices. This array must be at least as big as count. For every device handle in phDevices, this will provide the events that occurred for that device at the same position in this array. If no event was received for a given device, the corresponding array entry will be zero.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDriver
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == phDevices
nullptr == pNumDeviceEvents
nullptr == pEvents
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to listen to events.
ZE_RESULT_ERROR_INVALID_ARGUMENT
One or more of the supplied device handles belongs to a different driver.
zesDriverEventListenEx¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDriverEventListenEx
(ze_driver_handle_t hDriver, uint64_t timeout, uint32_t count, zes_device_handle_t *phDevices, uint32_t *pNumDeviceEvents, zes_event_type_flags_t *pEvents)¶ Wait for events to be received from a one or more devices.
- Parameters
hDriver
: [in] handle of the driver instancetimeout
: [in] if non-zero, then indicates the maximum time (in milliseconds) to yield before returning ZE_RESULT_SUCCESS or ZE_RESULT_NOT_READY; if zero, then will check status and return immediately; if UINT64_MAX, then function will not return until events arrive.count
: [in] Number of device handles in phDevices.phDevices
: [in][range(0, count)] Device handles to listen to for events. Only devices from the provided driver handle can be specified in this list.pNumDeviceEvents
: [in,out] Will contain the actual number of devices in phDevices that generated events. If non-zero, check pEvents to determine the devices and events that were received.pEvents
: [in,out] An array that will continue the list of events for each device listened in phDevices. This array must be at least as big as count. For every device handle in phDevices, this will provide the events that occurred for that device at the same position in this array. If no event was received for a given device, the corresponding array entry will be zero.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDriver
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == phDevices
nullptr == pNumDeviceEvents
nullptr == pEvents
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to listen to events.
ZE_RESULT_ERROR_INVALID_ARGUMENT
One or more of the supplied device handles belongs to a different driver.
Events Enums¶
zes_event_type_flags_t¶
-
enum
zes_event_type_flag_t
¶ Values:
-
ZES_EVENT_TYPE_FLAG_DEVICE_DETACH
= ZE_BIT(0)¶ Event is triggered when the device is no longer available (due to a reset or being disabled).
-
ZES_EVENT_TYPE_FLAG_DEVICE_ATTACH
= ZE_BIT(1)¶ Event is triggered after the device is available again.
-
ZES_EVENT_TYPE_FLAG_DEVICE_SLEEP_STATE_ENTER
= ZE_BIT(2)¶ deep sleep state
Event is triggered when the driver is about to put the device into a
-
ZES_EVENT_TYPE_FLAG_DEVICE_SLEEP_STATE_EXIT
= ZE_BIT(3)¶ sleep state
Event is triggered when the driver is waking the device up from a deep
-
ZES_EVENT_TYPE_FLAG_FREQ_THROTTLED
= ZE_BIT(4)¶ Event is triggered when the frequency starts being throttled.
-
ZES_EVENT_TYPE_FLAG_ENERGY_THRESHOLD_CROSSED
= ZE_BIT(5)¶ (use zesPowerSetEnergyThreshold() to configure).
Event is triggered when the energy consumption threshold is reached
-
ZES_EVENT_TYPE_FLAG_TEMP_CRITICAL
= ZE_BIT(6)¶ Event is triggered when the critical temperature is reached (use zesTemperatureSetConfig() to configure - disabled by default).
-
ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD1
= ZE_BIT(7)¶ Event is triggered when the temperature crosses threshold 1 (use zesTemperatureSetConfig() to configure - disabled by default).
-
ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD2
= ZE_BIT(8)¶ Event is triggered when the temperature crosses threshold 2 (use zesTemperatureSetConfig() to configure - disabled by default).
-
ZES_EVENT_TYPE_FLAG_MEM_HEALTH
= ZE_BIT(9)¶ Event is triggered when the health of device memory changes.
-
ZES_EVENT_TYPE_FLAG_FABRIC_PORT_HEALTH
= ZE_BIT(10)¶ Event is triggered when the health of fabric ports change.
-
ZES_EVENT_TYPE_FLAG_PCI_LINK_HEALTH
= ZE_BIT(11)¶ Event is triggered when the health of the PCI link changes.
-
ZES_EVENT_TYPE_FLAG_RAS_CORRECTABLE_ERRORS
= ZE_BIT(12)¶ Event is triggered when accelerator RAS correctable errors cross thresholds (use zesRasSetConfig() to configure - disabled by default).
-
ZES_EVENT_TYPE_FLAG_RAS_UNCORRECTABLE_ERRORS
= ZE_BIT(13)¶ Event is triggered when accelerator RAS uncorrectable errors cross thresholds (use zesRasSetConfig() to configure - disabled by default).
-
ZES_EVENT_TYPE_FLAG_DEVICE_RESET_REQUIRED
= ZE_BIT(14)¶ zesDeviceGetState() to determine the reasons for the reset).
Event is triggered when the device needs to be reset (use
-
ZES_EVENT_TYPE_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
Fabric¶
Functions
Enumerations
Structures
Fabric Functions¶
zesDeviceEnumFabricPorts¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumFabricPorts
(zes_device_handle_t hDevice, uint32_t *pCount, zes_fabric_port_handle_t *phPort)¶ Get handle of Fabric ports in a device.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phPort
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesFabricPortGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortGetProperties
(zes_fabric_port_handle_t hPort, zes_fabric_port_properties_t *pProperties)¶ Get Fabric port properties.
- Parameters
hPort
: [in] Handle for the component.pProperties
: [in,out] Will contain properties of the Fabric Port.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesFabricPortGetLinkType¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortGetLinkType
(zes_fabric_port_handle_t hPort, zes_fabric_link_type_t *pLinkType)¶ Get Fabric port link type.
- Parameters
hPort
: [in] Handle for the component.pLinkType
: [in,out] Will contain details about the link attached to the Fabric port.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pLinkType
zesFabricPortGetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortGetConfig
(zes_fabric_port_handle_t hPort, zes_fabric_port_config_t *pConfig)¶ Get Fabric port configuration.
- Parameters
hPort
: [in] Handle for the component.pConfig
: [in,out] Will contain configuration of the Fabric Port.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
zesFabricPortSetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortSetConfig
(zes_fabric_port_handle_t hPort, const zes_fabric_port_config_t *pConfig)¶ Set Fabric port configuration.
- Parameters
hPort
: [in] Handle for the component.pConfig
: [in] Contains new configuration of the Fabric Port.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFabricPortGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortGetState
(zes_fabric_port_handle_t hPort, zes_fabric_port_state_t *pState)¶ Get Fabric port state - status (health/degraded/failed/disabled), reasons for link degradation or instability, current rx/tx speed.
- Parameters
hPort
: [in] Handle for the component.pState
: [in,out] Will contain the current state of the Fabric Port
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesFabricPortGetThroughput¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFabricPortGetThroughput
(zes_fabric_port_handle_t hPort, zes_fabric_port_throughput_t *pThroughput)¶ Get Fabric port throughput.
- Parameters
hPort
: [in] Handle for the component.pThroughput
: [in,out] Will contain the Fabric port throughput counters.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPort
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pThroughput
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to query this telemetry.
Fabric Enums¶
zes_fabric_port_status_t¶
-
enum
zes_fabric_port_status_t
¶ Fabric port status.
Values:
-
ZES_FABRIC_PORT_STATUS_UNKNOWN
= 0¶ The port status cannot be determined.
-
ZES_FABRIC_PORT_STATUS_HEALTHY
= 1¶ The port is up and operating as expected.
-
ZES_FABRIC_PORT_STATUS_DEGRADED
= 2¶ The port is up but has quality and/or speed degradation.
-
ZES_FABRIC_PORT_STATUS_FAILED
= 3¶ Port connection instabilities are preventing workloads making forward progress
-
ZES_FABRIC_PORT_STATUS_DISABLED
= 4¶ The port is configured down.
-
ZES_FABRIC_PORT_STATUS_FORCE_UINT32
= 0x7fffffff¶
-
zes_fabric_port_qual_issue_flags_t¶
-
enum
zes_fabric_port_qual_issue_flag_t
¶ Values:
-
ZES_FABRIC_PORT_QUAL_ISSUE_FLAG_LINK_ERRORS
= ZE_BIT(0)¶ Excessive link errors are occurring.
-
ZES_FABRIC_PORT_QUAL_ISSUE_FLAG_SPEED
= ZE_BIT(1)¶ There is a degradation in the bitrate and/or width of the link.
-
ZES_FABRIC_PORT_QUAL_ISSUE_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
zes_fabric_port_failure_flags_t¶
-
enum
zes_fabric_port_failure_flag_t
¶ Values:
-
ZES_FABRIC_PORT_FAILURE_FLAG_FAILED
= ZE_BIT(0)¶ A previously operating link has failed. Hardware will automatically retrain this port. This state will persist until either the physical connection is removed or the link trains successfully.
-
ZES_FABRIC_PORT_FAILURE_FLAG_TRAINING_TIMEOUT
= ZE_BIT(1)¶ A connection has not been established within an expected time. Hardware will continue to attempt port training. This status will persist until either the physical connection is removed or the link successfully trains.
-
ZES_FABRIC_PORT_FAILURE_FLAG_FLAPPING
= ZE_BIT(2)¶ Port has excessively trained and then transitioned down for some period of time. Driver will allow port to continue to train, but will not enable the port for use until the port has been disabled and subsequently re-enabled using zesFabricPortSetConfig().
-
ZES_FABRIC_PORT_FAILURE_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
Fabric Structures¶
zes_fabric_port_id_t¶
-
struct
zes_fabric_port_id_t
¶ Unique identifier for a fabric port.
This not a universal identifier. The identified is garanteed to be unique for the current hardware configuration of the system. Changes in the hardware may result in a different identifier for a given port.
The main purpose of this identifier to build up an instantaneous topology map of system connectivity. An application should enumerate all fabric ports and match zes_fabric_port_state_t.remotePortId to zes_fabric_port_properties_t.portId.
zes_fabric_port_properties_t¶
-
struct
zes_fabric_port_properties_t
¶ Fabric port properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
char
model
[ZES_MAX_FABRIC_PORT_MODEL_SIZE
]¶ [out] Description of port technology. Will be set to the string “unkown” if this cannot be determined for this port.
-
ze_bool_t
onSubdevice
¶ [out] True if the port is located on a sub-device; false means that the port is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_fabric_port_id_t
portId
¶ [out] The unique port identifier
-
zes_fabric_port_speed_t
maxRxSpeed
¶ [out] Maximum speed supported by the receive side of the port (sum of all lanes)
-
zes_fabric_port_speed_t
maxTxSpeed
¶ [out] Maximum speed supported by the transmit side of the port (sum of all lanes)
-
zes_structure_type_t
zes_fabric_link_type_t¶
-
struct
zes_fabric_link_type_t
¶ Provides information about the fabric link attached to a port.
Public Members
-
char
desc
[ZES_MAX_FABRIC_LINK_TYPE_SIZE
]¶ [out] This provides a static textural description of the physic attachment type. Will be set to the string “unkown” if this cannot be determined for this port.
-
char
zes_fabric_port_config_t¶
-
struct
zes_fabric_port_config_t
¶ Fabric port configuration.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
ze_bool_t
enabled
¶ [in,out] Port is configured up/down
-
ze_bool_t
beaconing
¶ [in,out] Beaconing is configured on/off
-
zes_structure_type_t
zes_fabric_port_state_t¶
-
struct
zes_fabric_port_state_t
¶ Fabric port state.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_fabric_port_status_t
status
¶ [out] The current status of the port
-
zes_fabric_port_qual_issue_flags_t
qualityIssues
¶ [out] If status is ZES_FABRIC_PORT_STATUS_DEGRADED,
then this gives a combination of
zes_fabric_port_qual_issue_flag_t for quality issues that have been detected; otherwise, 0 indicates there are no quality issues with the link at this time.
-
zes_fabric_port_failure_flags_t
failureReasons
¶ [out] If status is ZES_FABRIC_PORT_STATUS_FAILED, then this gives a combination of zes_fabric_port_failure_flag_t for reasons for the connection instability; otherwise, 0 indicates there are no connection stability issues at this time.
-
zes_fabric_port_id_t
remotePortId
¶ [out] The unique port identifier for the remote connection point if status is ZES_FABRIC_PORT_STATUS_HEALTHY, ZES_FABRIC_PORT_STATUS_DEGRADED or ZES_FABRIC_PORT_STATUS_FAILED
-
zes_fabric_port_speed_t
rxSpeed
¶ [out] Current maximum receive speed (sum of all lanes)
-
zes_fabric_port_speed_t
txSpeed
¶ [out] Current maximum transmit speed (sum of all lanes)
-
zes_structure_type_t
zes_fabric_port_throughput_t¶
-
struct
zes_fabric_port_throughput_t
¶ Fabric port throughput.
Public Members
-
uint64_t
timestamp
¶ [out] Monotonic timestamp counter in microseconds when the measurement was made. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
-
uint64_t
rxCounter
¶ [out] Monotonic counter for the number of bytes received (sum of all lanes). This includes all protocol overhead, not only the GPU traffic.
-
uint64_t
txCounter
¶ [out] Monotonic counter for the number of bytes transmitted (sum of all lanes). This includes all protocol overhead, not only the GPU traffic.
-
uint64_t
Fan¶
Functions
Enumerations
Structures
Fan Functions¶
zesDeviceEnumFans¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumFans
(zes_device_handle_t hDevice, uint32_t *pCount, zes_fan_handle_t *phFan)¶ Get handle of fans.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phFan
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesFanGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanGetProperties
(zes_fan_handle_t hFan, zes_fan_properties_t *pProperties)¶ Get fan properties.
- Parameters
hFan
: [in] Handle for the component.pProperties
: [in,out] Will contain the properties of the fan.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesFanGetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanGetConfig
(zes_fan_handle_t hFan, zes_fan_config_t *pConfig)¶ Get fan configurations and the current fan speed mode (default, fixed, temp-speed table)
- Parameters
hFan
: [in] Handle for the component.pConfig
: [in,out] Will contain the current configuration of the fan.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
zesFanSetDefaultMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanSetDefaultMode
(zes_fan_handle_t hFan)¶ Configure the fan to run with hardware factory settings (set mode to ZES_FAN_SPEED_MODE_DEFAULT)
- Parameters
hFan
: [in] Handle for the component.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFanSetFixedSpeedMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanSetFixedSpeedMode
(zes_fan_handle_t hFan, const zes_fan_speed_t *speed)¶ Configure the fan to rotate at a fixed speed (set mode to ZES_FAN_SPEED_MODE_FIXED)
- Parameters
hFan
: [in] Handle for the component.speed
: [in] The fixed fan speed setting
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == speed
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Fixing the fan speed not supported by the hardware or the fan speed units are not supported. See zes_fan_properties_t.supportedModes and zes_fan_properties_t.supportedUnits.
zesFanSetSpeedTableMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanSetSpeedTableMode
(zes_fan_handle_t hFan, const zes_fan_speed_table_t *speedTable)¶ Configure the fan to adjust speed based on a temperature/speed table (set mode to ZES_FAN_SPEED_MODE_TABLE)
- Parameters
hFan
: [in] Handle for the component.speedTable
: [in] A table containing temperature/speed pairs.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == speedTable
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
ZE_RESULT_ERROR_INVALID_ARGUMENT
The temperature/speed pairs in the array are not sorted on temperature from lowest to highest.
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Fan speed table not supported by the hardware or the fan speed units are not supported. See zes_fan_properties_t.supportedModes and zes_fan_properties_t.supportedUnits.
zesFanGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFanGetState
(zes_fan_handle_t hFan, zes_fan_speed_units_t units, int32_t *pSpeed)¶ Get current state of a fan - current mode and speed.
- Parameters
hFan
: [in] Handle for the component.units
: [in] The units in which the fan speed should be returned.pSpeed
: [in,out] Will contain the current speed of the fan in the units requested. A value of -1 indicates that the fan speed cannot be measured.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFan
ZE_RESULT_ERROR_INVALID_ENUMERATION
ZES_FAN_SPEED_UNITS_PERCENT < units
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pSpeed
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
The requested fan speed units are not supported. See zes_fan_properties_t.supportedUnits.
Fan Enums¶
zes_fan_speed_mode_t¶
-
enum
zes_fan_speed_mode_t
¶ Fan resource speed mode.
Values:
-
ZES_FAN_SPEED_MODE_DEFAULT
= 0¶ The fan speed is operating using the hardware default settings.
-
ZES_FAN_SPEED_MODE_FIXED
= 1¶ The fan speed is currently set to a fixed value.
-
ZES_FAN_SPEED_MODE_TABLE
= 2¶ The fan speed is currently controlled dynamically by hardware based on a temp/speed table
-
ZES_FAN_SPEED_MODE_FORCE_UINT32
= 0x7fffffff¶
-
zes_fan_speed_units_t¶
Fan Structures¶
zes_fan_speed_t¶
-
struct
zes_fan_speed_t
¶ Fan speed.
Public Members
-
int32_t
speed
¶ [in,out] The speed of the fan. On output, a value of -1 indicates that there is no fixed fan speed setting.
-
zes_fan_speed_units_t
units
¶ [in,out] The units that the fan speed is expressed in. On output, if fan speed is -1 then units should be ignored.
-
int32_t
zes_fan_temp_speed_t¶
-
struct
zes_fan_temp_speed_t
¶ Fan temperature/speed pair.
Public Members
-
uint32_t
temperature
¶ [in,out] Temperature in degrees Celsius.
-
zes_fan_speed_t
speed
¶ [in,out] The speed of the fan
-
uint32_t
zes_fan_speed_table_t¶
-
struct
zes_fan_speed_table_t
¶ Fan speed table.
Public Members
-
int32_t
numPoints
¶ [in,out] The number of valid points in the fan speed table. 0 means that there is no fan speed table configured. -1 means that a fan speed table is not supported by the hardware.
-
zes_fan_temp_speed_t
table
[ZES_FAN_TEMP_SPEED_PAIR_COUNT
]¶ based on temperature from lowest to highest.
[in,out] Array of temperature/fan speed pairs. The table is ordered
-
int32_t
zes_fan_properties_t¶
-
struct
zes_fan_properties_t
¶ Fan properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Indicates if software can control the fan speed assuming the user has permissions
-
uint32_t
supportedModes
¶ [out] Bitfield of supported fan configuration modes (1<<zes_fan_speed_mode_t)
-
uint32_t
supportedUnits
¶ [out] Bitfield of supported fan speed units (1<<zes_fan_speed_units_t)
-
int32_t
maxRPM
¶ [out] The maximum RPM of the fan. A value of -1 means that this property is unknown.
-
int32_t
maxPoints
¶ [out] The maximum number of points in the fan temp/speed table. A value of -1 means that this fan doesn’t support providing a temp/speed table.
-
zes_structure_type_t
zes_fan_config_t¶
-
struct
zes_fan_config_t
¶ Fan configuration.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_fan_speed_mode_t
mode
¶ [in,out] The fan speed mode (fixed, temp-speed table)
-
zes_fan_speed_t
speedFixed
¶ [in,out] The current fixed fan speed setting
-
zes_fan_speed_table_t
speedTable
¶ [out] A table containing temperature/speed pairs
-
zes_structure_type_t
Firmware¶
Functions
Structures
Firmware Functions¶
zesDeviceEnumFirmwares¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumFirmwares
(zes_device_handle_t hDevice, uint32_t *pCount, zes_firmware_handle_t *phFirmware)¶ Get handle of firmwares.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phFirmware
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesFirmwareGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFirmwareGetProperties
(zes_firmware_handle_t hFirmware, zes_firmware_properties_t *pProperties)¶ Get firmware properties.
- Parameters
hFirmware
: [in] Handle for the component.pProperties
: [in,out] Pointer to an array that will hold the properties of the firmware
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFirmware
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesFirmwareFlash¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFirmwareFlash
(zes_firmware_handle_t hFirmware, void *pImage, uint32_t size)¶ Flash a new firmware image.
- Parameters
hFirmware
: [in] Handle for the component.pImage
: [in] Image of the new firmware to flash.size
: [in] Size of the flash image.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFirmware
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pImage
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to perform this operation.
Firmware Structures¶
zes_firmware_properties_t¶
-
struct
zes_firmware_properties_t
¶ Firmware properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Indicates if software can flash the firmware assuming the user has permissions
-
char
name
[ZES_STRING_PROPERTY_SIZE
]¶ [out] NULL terminated string value. The string “unknown” will be returned if this property cannot be determined.
-
char
version
[ZES_STRING_PROPERTY_SIZE
]¶ [out] NULL terminated string value. The string “unknown” will be returned if this property cannot be determined.
-
zes_structure_type_t
Frequency¶
Functions
Enumerations
Structures
Frequency Functions¶
zesDeviceEnumFrequencyDomains¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumFrequencyDomains
(zes_device_handle_t hDevice, uint32_t *pCount, zes_freq_handle_t *phFrequency)¶ Get handle of frequency domains.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phFrequency
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesFrequencyGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyGetProperties
(zes_freq_handle_t hFrequency, zes_freq_properties_t *pProperties)¶ Get frequency properties - available frequencies.
- Parameters
hFrequency
: [in] Handle for the component.pProperties
: [in,out] The frequency properties for the specified domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesFrequencyGetAvailableClocks¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyGetAvailableClocks
(zes_freq_handle_t hFrequency, uint32_t *pCount, double *phFrequency)¶ Get available non-overclocked hardware clock frequencies for the frequency domain.
- Parameters
hFrequency
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of frequencies. if count is zero, then the driver shall update the value with the total number of frequencies that are available. if count is greater than the number of frequencies that are available, then the driver shall update the value with the correct number of frequencies.phFrequency
: [in,out][optional][range(0, *pCount)] array of frequencies in units of MHz and sorted from slowest to fastest. if count is less than the number of frequencies that are available, then the driver shall only retrieve that number of frequencies.
The list of available frequencies is returned in order of slowest to fastest.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesFrequencyGetRange¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyGetRange
(zes_freq_handle_t hFrequency, zes_freq_range_t *pLimits)¶ Get current frequency limits.
- Parameters
hFrequency
: [in] Handle for the component.pLimits
: [in,out] The range between which the hardware can operate for the specified domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pLimits
zesFrequencySetRange¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencySetRange
(zes_freq_handle_t hFrequency, const zes_freq_range_t *pLimits)¶ Set frequency range between which the hardware can operate.
- Parameters
hFrequency
: [in] Handle for the component.pLimits
: [in] The limits between which the hardware can operate for the specified domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pLimits
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyGetState
(zes_freq_handle_t hFrequency, zes_freq_state_t *pState)¶ Get current frequency state - frequency request, actual frequency, TDP limits.
- Parameters
hFrequency
: [in] Handle for the component.pState
: [in,out] Frequency state for the specified domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesFrequencyGetThrottleTime¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyGetThrottleTime
(zes_freq_handle_t hFrequency, zes_freq_throttle_time_t *pThrottleTime)¶ Get frequency throttle time.
- Parameters
hFrequency
: [in] Handle for the component.pThrottleTime
: [in,out] Will contain a snapshot of the throttle time counters for the specified domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pThrottleTime
zesFrequencyOcGetCapabilities¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetCapabilities
(zes_freq_handle_t hFrequency, zes_oc_capabilities_t *pOcCapabilities)¶ Get the overclocking capabilities.
- Parameters
hFrequency
: [in] Handle for the component.pOcCapabilities
: [in,out] Pointer to the capabilities structure zes_oc_capabilities_t.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pOcCapabilities
zesFrequencyOcGetFrequencyTarget¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetFrequencyTarget
(zes_freq_handle_t hFrequency, double *pCurrentOcFrequency)¶ Get the current overclocking frequency target, if extended moded is supported, will returned in 1 Mhz granularity.
- Parameters
hFrequency
: [in] Handle for the component.pCurrentOcFrequency
: [out] Overclocking Frequency in MHz, if extended moded is supported, will returned in 1 Mhz granularity, else, in multiples of 50 Mhz. This cannot be greater than zes_oc_capabilities_t.maxOcFrequency.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCurrentOcFrequency
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcSetFrequencyTarget¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcSetFrequencyTarget
(zes_freq_handle_t hFrequency, double CurrentOcFrequency)¶ Set the current overclocking frequency target, if extended moded is supported, can be set in 1 Mhz granularity.
- Parameters
hFrequency
: [in] Handle for the component.CurrentOcFrequency
: [in] Overclocking Frequency in MHz, if extended moded is supported, it could be set in 1 Mhz granularity, else, in multiples of 50 Mhz. This cannot be greater than zes_oc_capabilities_t.maxOcFrequency.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcGetVoltageTarget¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetVoltageTarget
(zes_freq_handle_t hFrequency, double *pCurrentVoltageTarget, double *pCurrentVoltageOffset)¶ Get the current overclocking voltage settings.
- Parameters
hFrequency
: [in] Handle for the component.pCurrentVoltageTarget
: [out] Overclock voltage in Volts. This cannot be greater than zes_oc_capabilities_t.maxOcVoltage.pCurrentVoltageOffset
: [out] This voltage offset is applied to all points on the voltage/frequency curve, include the new overclock voltageTarget. It can be in the range (zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCurrentVoltageTarget
nullptr == pCurrentVoltageOffset
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcSetVoltageTarget¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcSetVoltageTarget
(zes_freq_handle_t hFrequency, double CurrentVoltageTarget, double CurrentVoltageOffset)¶ Set the current overclocking voltage settings.
- Parameters
hFrequency
: [in] Handle for the component.CurrentVoltageTarget
: [in] Overclock voltage in Volts. This cannot be greater than zes_oc_capabilities_t.maxOcVoltage.CurrentVoltageOffset
: [in] This voltage offset is applied to all points on the voltage/frequency curve, include the new overclock voltageTarget. It can be in the range (zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcSetMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcSetMode
(zes_freq_handle_t hFrequency, zes_oc_mode_t CurrentOcMode)¶ Set the current overclocking mode.
- Parameters
hFrequency
: [in] Handle for the component.CurrentOcMode
: [in] Current Overclocking Mode zes_oc_mode_t.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_ENUMERATION
ZES_OC_MODE_FIXED < CurrentOcMode
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcGetMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetMode
(zes_freq_handle_t hFrequency, zes_oc_mode_t *pCurrentOcMode)¶ Get the current overclocking mode.
- Parameters
hFrequency
: [in] Handle for the component.pCurrentOcMode
: [out] Current Overclocking Mode zes_oc_mode_t.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCurrentOcMode
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
The specified voltage and/or frequency overclock settings exceed the hardware values (see zes_oc_capabilities_t.maxOcFrequency, zes_oc_capabilities_t.maxOcVoltage, zes_oc_capabilities_t.minOcVoltageOffset, zes_oc_capabilities_t.maxOcVoltageOffset).
Requested voltage overclock is very high but zes_oc_capabilities_t.isHighVoltModeEnabled is not enabled for the device.
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcGetIccMax¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetIccMax
(zes_freq_handle_t hFrequency, double *pOcIccMax)¶ Get the maximum current limit setting.
- Parameters
hFrequency
: [in] Handle for the component.pOcIccMax
: [in,out] Will contain the maximum current limit in Amperes on successful return.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pOcIccMax
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
Capability zes_oc_capabilities_t.isIccMaxSupported is false for this frequency domain
zesFrequencyOcSetIccMax¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcSetIccMax
(zes_freq_handle_t hFrequency, double ocIccMax)¶ Change the maximum current limit setting.
- Parameters
hFrequency
: [in] Handle for the component.ocIccMax
: [in] The new maximum current limit in Amperes.
Setting ocIccMax to 0.0 will return the value to the factory default.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
Capability zes_oc_capabilities_t.isIccMaxSupported is false for this frequency domain
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INVALID_ARGUMENT
The specified current limit is too low or too high
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesFrequencyOcGetTjMax¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcGetTjMax
(zes_freq_handle_t hFrequency, double *pOcTjMax)¶ Get the maximum temperature limit setting.
- Parameters
hFrequency
: [in] Handle for the component.pOcTjMax
: [in,out] Will contain the maximum temperature limit in degrees Celsius on successful return.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pOcTjMax
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
zesFrequencyOcSetTjMax¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesFrequencyOcSetTjMax
(zes_freq_handle_t hFrequency, double ocTjMax)¶ Change the maximum temperature limit setting.
- Parameters
hFrequency
: [in] Handle for the component.ocTjMax
: [in] The new maximum temperature limit in degrees Celsius.
Setting ocTjMax to 0.0 will return the value to the factory default.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hFrequency
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Overclocking is not supported on this frequency domain (zes_oc_capabilities_t.isOcSupported)
Capability zes_oc_capabilities_t.isTjMaxSupported is false for this frequency domain
Overclocking feature is locked on this frequency domain
ZE_RESULT_ERROR_INVALID_ARGUMENT
The specified temperature limit is too high
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
Frequency Enums¶
zes_freq_domain_t¶
zes_freq_throttle_reason_flags_t¶
-
enum
zes_freq_throttle_reason_flag_t
¶ Values:
-
ZES_FREQ_THROTTLE_REASON_FLAG_AVE_PWR_CAP
= ZE_BIT(0)¶ frequency throttled due to average power excursion (PL1)
-
ZES_FREQ_THROTTLE_REASON_FLAG_BURST_PWR_CAP
= ZE_BIT(1)¶ frequency throttled due to burst power excursion (PL2)
-
ZES_FREQ_THROTTLE_REASON_FLAG_CURRENT_LIMIT
= ZE_BIT(2)¶ frequency throttled due to current excursion (PL4)
-
ZES_FREQ_THROTTLE_REASON_FLAG_THERMAL_LIMIT
= ZE_BIT(3)¶ frequency throttled due to thermal excursion (T > TjMax)
-
ZES_FREQ_THROTTLE_REASON_FLAG_PSU_ALERT
= ZE_BIT(4)¶ frequency throttled due to power supply assertion
-
ZES_FREQ_THROTTLE_REASON_FLAG_SW_RANGE
= ZE_BIT(5)¶ frequency throttled due to software supplied frequency range
-
ZES_FREQ_THROTTLE_REASON_FLAG_HW_RANGE
= ZE_BIT(6)¶ range when it receives clocks
frequency throttled due to a sub block that has a lower frequency
-
ZES_FREQ_THROTTLE_REASON_FLAG_FORCE_UINT32
= 0x7fffffff¶
-
zes_oc_mode_t¶
-
enum
zes_oc_mode_t
¶ Overclocking modes.
Values:
-
ZES_OC_MODE_OFF
= 0¶ Overclocking if off - hardware is running using factory default voltages/frequencies.
-
ZES_OC_MODE_OVERRIDE
= 1¶ Overclock override mode - In this mode, a fixed user-supplied voltage is applied independent of the frequency request. The maximum permitted frequency can also be increased. This mode disables INTERPOLATIVE and FIXED modes.
-
ZES_OC_MODE_INTERPOLATIVE
= 2¶ Overclock interpolative mode - In this mode, the voltage/frequency curve can be extended with a new voltage/frequency point that will be interpolated. The existing voltage/frequency points can also be offset (up or down) by a fixed voltage. This mode disables FIXED and OVERRIDE modes.
-
ZES_OC_MODE_FIXED
= 3¶ Overclocking fixed Mode - In this mode, hardware will disable most frequency throttling and lock the frequency and voltage at the specified overclock values. This mode disables OVERRIDE and INTERPOLATIVE modes. This mode can damage the part, most of the protections are disabled on this mode.
-
ZES_OC_MODE_FORCE_UINT32
= 0x7fffffff¶
-
Frequency Structures¶
zes_freq_properties_t¶
-
struct
zes_freq_properties_t
¶ Frequency properties.
Indicates if this frequency domain can be overclocked (if true, functions such as zesFrequencyOcSetFrequencyTarget() are supported).
The min/max hardware frequencies are specified for non-overclock configurations. For overclock configurations, use zesFrequencyOcGetFrequencyTarget() to determine the maximum frequency that can be requested.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_freq_domain_t
type
¶ [out] The hardware block that this frequency domain controls (GPU, memory, …)
-
ze_bool_t
onSubdevice
¶ [out] True if this resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Indicates if software can control the frequency of this domain assuming the user has permissions
-
ze_bool_t
isThrottleEventSupported
¶ [out] Indicates if software can register to receive event ZES_EVENT_TYPE_FLAG_FREQ_THROTTLED
-
double
min
¶ [out] The minimum hardware clock frequency in units of MHz.
-
double
max
¶ [out] The maximum non-overclock hardware clock frequency in units of MHz.
zes_freq_range_t¶
-
struct
zes_freq_range_t
¶ Frequency range between which the hardware can operate. The limits can be above or below the hardware limits - the hardware will clamp appropriately.
Public Members
-
double
min
¶ [in,out] The min frequency in MHz below which hardware frequency management will not request frequencies. On input, setting to 0 will permit the frequency to go down to the hardware minimum. On output, a negative value indicates that no external minimum frequency limit is in effect.
-
double
max
¶ [in,out] The max frequency in MHz above which hardware frequency management will not request frequencies. On input, setting to 0 or a very big number will permit the frequency to go all the way up to the hardware maximum. On output, a negative number indicates that no external maximum frequency limit is in effect.
-
double
zes_freq_state_t¶
-
struct
zes_freq_state_t
¶ Frequency state.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
double
currentVoltage
¶ [out] Current voltage in Volts. A negative value indicates that this property is not known.
-
double
request
¶ [out] The current frequency request in MHz. A negative value indicates that this property is not known.
-
double
tdp
¶ [out] The maximum frequency in MHz supported under the current TDP conditions. This fluctuates dynamically based on the power and thermal limits of the part. A negative value indicates that this property is not known.
-
double
efficient
¶ [out] The efficient minimum frequency in MHz. A negative value indicates that this property is not known.
-
double
actual
¶ [out] The resolved frequency in MHz. A negative value indicates that this property is not known.
-
zes_freq_throttle_reason_flags_t
throttleReasons
¶ Returns 0 (frequency not throttled) or a combination of zes_freq_throttle_reason_flag_t.
[out] The reasons that the frequency is being limited by the hardware.
-
zes_structure_type_t
zes_freq_throttle_time_t¶
-
struct
zes_freq_throttle_time_t
¶ Frequency throttle time snapshot.
Percent time throttled is calculated by taking two snapshots (s1, s2) and using the equation: throttled = (s2.throttleTime - s1.throttleTime) / (s2.timestamp - s1.timestamp)
Public Members
-
uint64_t
throttleTime
¶ [out] The monotonic counter of time in microseconds that the frequency has been limited by the hardware.
-
uint64_t
timestamp
¶ [out] Microsecond timestamp when throttleTime was captured. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
zes_oc_capabilities_t¶
-
struct
zes_oc_capabilities_t
¶ Overclocking properties.
Provides all the overclocking capabilities and properties supported by the device for the frequency domain.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
ze_bool_t
isOcSupported
¶ [out] Indicates if any overclocking features are supported on this frequency domain.
-
double
maxFactoryDefaultFrequency
¶ [out] Factory default non-overclock maximum frequency in Mhz.
-
double
maxFactoryDefaultVoltage
¶ [out] Factory default voltage used for the non-overclock maximum frequency in MHz.
-
double
maxOcFrequency
¶ [out] Maximum hardware overclocking frequency limit in Mhz.
-
double
minOcVoltageOffset
¶ [out] The minimum voltage offset that can be applied to the voltage/frequency curve. Note that this number can be negative.
-
double
maxOcVoltageOffset
¶ [out] The maximum voltage offset that can be applied to the voltage/frequency curve.
-
double
maxOcVoltage
¶ [out] The maximum overclock voltage that hardware supports.
-
ze_bool_t
isTjMaxSupported
¶ [out] Indicates if the maximum temperature limit (TjMax) can be changed for this frequency domain.
-
ze_bool_t
isIccMaxSupported
¶ [out] Indicates if the maximum current (IccMax) can be changed for this frequency domain.
-
ze_bool_t
isHighVoltModeCapable
¶ [out] Indicates if this frequency domains supports a feature to set very high voltages.
-
ze_bool_t
isHighVoltModeEnabled
¶ [out] Indicates if very high voltages are permitted on this frequency domain.
-
ze_bool_t
isExtendedModeSupported
¶ [out] Indicates if the extended overclocking features are supported. If this is supported, increments are on 1 Mhz basis.
-
ze_bool_t
isFixedModeSupported
¶ [out] Indicates if the fixed mode is supported. In this mode, hardware will disable most frequency throttling and lock the frequency and voltage at the specified overclock values.
Led¶
Functions
Structures
Led Functions¶
zesDeviceEnumLeds¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumLeds
(zes_device_handle_t hDevice, uint32_t *pCount, zes_led_handle_t *phLed)¶ Get handle of LEDs.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phLed
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesLedGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesLedGetProperties
(zes_led_handle_t hLed, zes_led_properties_t *pProperties)¶ Get LED properties.
- Parameters
hLed
: [in] Handle for the component.pProperties
: [in,out] Will contain the properties of the LED.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hLed
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesLedGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesLedGetState
(zes_led_handle_t hLed, zes_led_state_t *pState)¶ Get current state of a LED - on/off, color.
- Parameters
hLed
: [in] Handle for the component.pState
: [in,out] Will contain the current state of the LED.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hLed
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesLedSetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesLedSetState
(zes_led_handle_t hLed, ze_bool_t enable)¶ Turn the LED on/off.
- Parameters
hLed
: [in] Handle for the component.enable
: [in] Set to TRUE to turn the LED on, FALSE to turn off.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hLed
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
zesLedSetColor¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesLedSetColor
(zes_led_handle_t hLed, const zes_led_color_t *pColor)¶ Set the color of the LED.
- Parameters
hLed
: [in] Handle for the component.pColor
: [in] New color of the LED.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hLed
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pColor
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This LED doesn’t not support color changes. See zes_led_properties_t.haveRGB.
Led Structures¶
zes_led_properties_t¶
-
struct
zes_led_properties_t
¶ LED properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Indicates if software can control the LED assuming the user has permissions
-
ze_bool_t
haveRGB
¶ [out] Indicates if the LED is RGB capable
-
zes_structure_type_t
zes_led_color_t¶
-
struct
zes_led_color_t
¶ LED color.
Public Members
-
double
red
¶ [in,out][range(0.0, 1.0)] The LED red value. On output, a value less than 0.0 indicates that the color is not known.
-
double
green
¶ [in,out][range(0.0, 1.0)] The LED green value. On output, a value less than 0.0 indicates that the color is not known.
-
double
blue
¶ [in,out][range(0.0, 1.0)] The LED blue value. On output, a value less than 0.0 indicates that the color is not known.
-
double
zes_led_state_t¶
-
struct
zes_led_state_t
¶ LED state.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
ze_bool_t
isOn
¶ [out] Indicates if the LED is on or off
-
zes_led_color_t
color
¶ [out] Color of the LED
-
zes_structure_type_t
Memory¶
Functions
Enumerations
Structures
Memory Functions¶
zesDeviceEnumMemoryModules¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumMemoryModules
(zes_device_handle_t hDevice, uint32_t *pCount, zes_mem_handle_t *phMemory)¶ Get handle of memory modules.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phMemory
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesMemoryGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesMemoryGetProperties
(zes_mem_handle_t hMemory, zes_mem_properties_t *pProperties)¶ Get memory properties.
- Parameters
hMemory
: [in] Handle for the component.pProperties
: [in,out] Will contain memory properties.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hMemory
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesMemoryGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesMemoryGetState
(zes_mem_handle_t hMemory, zes_mem_state_t *pState)¶ Get memory state - health, allocated.
- Parameters
hMemory
: [in] Handle for the component.pState
: [in,out] Will contain the current health and allocated memory.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hMemory
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
zesMemoryGetBandwidth¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesMemoryGetBandwidth
(zes_mem_handle_t hMemory, zes_mem_bandwidth_t *pBandwidth)¶ Get memory bandwidth.
- Parameters
hMemory
: [in] Handle for the component.pBandwidth
: [in,out] Will contain the current health, free memory, total memory size.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hMemory
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pBandwidth
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to query this telemetry.
Memory Enums¶
zes_mem_type_t¶
-
enum
zes_mem_type_t
¶ Memory module types.
Values:
-
ZES_MEM_TYPE_HBM
= 0¶ HBM memory.
-
ZES_MEM_TYPE_DDR
= 1¶ DDR memory.
-
ZES_MEM_TYPE_DDR3
= 2¶ DDR3 memory.
-
ZES_MEM_TYPE_DDR4
= 3¶ DDR4 memory.
-
ZES_MEM_TYPE_DDR5
= 4¶ DDR5 memory.
-
ZES_MEM_TYPE_LPDDR
= 5¶ LPDDR memory.
-
ZES_MEM_TYPE_LPDDR3
= 6¶ LPDDR3 memory.
-
ZES_MEM_TYPE_LPDDR4
= 7¶ LPDDR4 memory.
-
ZES_MEM_TYPE_LPDDR5
= 8¶ LPDDR5 memory.
-
ZES_MEM_TYPE_SRAM
= 9¶ SRAM memory.
-
ZES_MEM_TYPE_L1
= 10¶ L1 cache.
-
ZES_MEM_TYPE_L3
= 11¶ L3 cache.
-
ZES_MEM_TYPE_GRF
= 12¶ Execution unit register file.
-
ZES_MEM_TYPE_SLM
= 13¶ Execution unit shared local memory.
-
ZES_MEM_TYPE_GDDR4
= 14¶ GDDR4 memory.
-
ZES_MEM_TYPE_GDDR5
= 15¶ GDDR5 memory.
-
ZES_MEM_TYPE_GDDR5X
= 16¶ GDDR5X memory.
-
ZES_MEM_TYPE_GDDR6
= 17¶ GDDR6 memory.
-
ZES_MEM_TYPE_GDDR6X
= 18¶ GDDR6X memory.
-
ZES_MEM_TYPE_GDDR7
= 19¶ GDDR7 memory.
-
ZES_MEM_TYPE_FORCE_UINT32
= 0x7fffffff¶
-
zes_mem_loc_t¶
zes_mem_health_t¶
-
enum
zes_mem_health_t
¶ Memory health.
Values:
-
ZES_MEM_HEALTH_UNKNOWN
= 0¶ The memory health cannot be determined.
-
ZES_MEM_HEALTH_OK
= 1¶ All memory channels are healthy.
-
ZES_MEM_HEALTH_DEGRADED
= 2¶ Excessive correctable errors have been detected on one or more channels. Device should be reset.
-
ZES_MEM_HEALTH_CRITICAL
= 3¶ Operating with reduced memory to cover banks with too many uncorrectable errors.
-
ZES_MEM_HEALTH_REPLACE
= 4¶ Device should be replaced due to excessive uncorrectable errors.
-
ZES_MEM_HEALTH_FORCE_UINT32
= 0x7fffffff¶
-
Memory Structures¶
zes_mem_properties_t¶
-
struct
zes_mem_properties_t
¶ Memory properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_mem_type_t
type
¶ [out] The memory type
-
ze_bool_t
onSubdevice
¶ [out] True if this resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_mem_loc_t
location
¶ [out] Location of this memory (system, device)
-
uint64_t
physicalSize
¶ [out] Physical memory size in bytes. A value of 0 indicates that this property is not known. However, a call to zesMemoryGetState() will correctly return the total size of usable memory.
-
int32_t
busWidth
¶ [out] Width of the memory bus. A value of -1 means that this property is unknown.
-
int32_t
numChannels
¶ [out] The number of memory channels. A value of -1 means that this property is unknown.
-
zes_structure_type_t
zes_mem_state_t¶
-
struct
zes_mem_state_t
¶ Memory state - health, allocated.
Percent allocation is given by 100 * (size - free / size.
Percent free is given by 100 * free / size.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_mem_health_t
health
¶ [out] Indicates the health of the memory
-
uint64_t
free
¶ [out] The free memory in bytes
-
uint64_t
size
¶ [out] The total allocatable memory in bytes (can be less than zes_mem_properties_t.physicalSize)
zes_mem_bandwidth_t¶
-
struct
zes_mem_bandwidth_t
¶ Memory bandwidth.
Percent bandwidth is calculated by taking two snapshots (s1, s2) and using the equation: bw = 10^6 * ((s2.readCounter - s1.readCounter) + (s2.writeCounter - s1.writeCounter)) / (s2.maxBandwidth * (s2.timestamp - s1.timestamp))
Public Members
-
uint64_t
readCounter
¶ [out] Total bytes read from memory
-
uint64_t
writeCounter
¶ [out] Total bytes written to memory
-
uint64_t
maxBandwidth
¶ [out] Current maximum bandwidth in units of bytes/sec
-
uint64_t
timestamp
¶ [out] The timestamp when these measurements were sampled. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
Performance¶
Functions
Structures
Performance Functions¶
zesDeviceEnumPerformanceFactorDomains¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumPerformanceFactorDomains
(zes_device_handle_t hDevice, uint32_t *pCount, zes_perf_handle_t *phPerf)¶ Get handles to accelerator domains whose performance can be optimized via a Performance Factor.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phPerf
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
A Performance Factor should be tuned for each workload.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesPerformanceFactorGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPerformanceFactorGetProperties
(zes_perf_handle_t hPerf, zes_perf_properties_t *pProperties)¶ Get properties about a Performance Factor domain.
- Parameters
hPerf
: [in] Handle for the Performance Factor domain.pProperties
: [in,out] Will contain information about the specified Performance Factor domain.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPerf
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesPerformanceFactorGetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPerformanceFactorGetConfig
(zes_perf_handle_t hPerf, double *pFactor)¶ Get current Performance Factor for a given domain.
- Parameters
hPerf
: [in] Handle for the Performance Factor domain.pFactor
: [in,out] Will contain the actual Performance Factor being used by the hardware (may not be the same as the requested Performance Factor).
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPerf
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pFactor
zesPerformanceFactorSetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPerformanceFactorSetConfig
(zes_perf_handle_t hPerf, double factor)¶ Change the performance factor for a domain.
- Parameters
hPerf
: [in] Handle for the Performance Factor domain.factor
: [in] The new Performance Factor.
The Performance Factor is a number between 0 and 100.
A Performance Factor is a hint to the hardware. Depending on the hardware, the request may not be granted. Follow up this function with a call to zesPerformanceFactorGetConfig() to determine the actual factor being used by the hardware.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
Performance Structures¶
zes_perf_properties_t¶
-
struct
zes_perf_properties_t
¶ Static information about a Performance Factor domain.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if this Performance Factor affects accelerators located on a sub-device
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_engine_type_flags_t
engines
¶ [out] Bitfield of accelerator engine types that are affected by this Performance Factor.
-
zes_structure_type_t
Power¶
Functions
Structures
Power Functions¶
zesDeviceEnumPowerDomains¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumPowerDomains
(zes_device_handle_t hDevice, uint32_t *pCount, zes_pwr_handle_t *phPower)¶ Get handle of power domains.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phPower
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesDeviceGetCardPowerDomain¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceGetCardPowerDomain
(zes_device_handle_t hDevice, zes_pwr_handle_t *phPower)¶ Get handle of the PCIe card-level power.
- Parameters
hDevice
: [in] Sysman handle of the device.phPower
: [in,out] power domain handle for the entire PCIe card.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == phPower
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
The device does not provide access to card level power controls or telemetry. An invalid power domain handle will be returned in phPower.
zesPowerGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerGetProperties
(zes_pwr_handle_t hPower, zes_power_properties_t *pProperties)¶ Get properties related to a power domain.
- Parameters
hPower
: [in] Handle for the component.pProperties
: [in,out] Structure that will contain property data.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPower
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesPowerGetEnergyCounter¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerGetEnergyCounter
(zes_pwr_handle_t hPower, zes_power_energy_counter_t *pEnergy)¶ Get energy counter.
- Parameters
hPower
: [in] Handle for the component.pEnergy
: [in,out] Will contain the latest snapshot of the energy counter and timestamp when the last counter value was measured.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPower
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pEnergy
zesPowerGetLimits¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerGetLimits
(zes_pwr_handle_t hPower, zes_power_sustained_limit_t *pSustained, zes_power_burst_limit_t *pBurst, zes_power_peak_limit_t *pPeak)¶ Get power limits.
- Parameters
hPower
: [in] Handle for the component.pSustained
: [in,out][optional] The sustained power limit. If this is null, the current sustained power limits will not be returned.pBurst
: [in,out][optional] The burst power limit. If this is null, the current peak power limits will not be returned.pPeak
: [in,out][optional] The peak power limit. If this is null, the peak power limits will not be returned.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
zesPowerSetLimits¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerSetLimits
(zes_pwr_handle_t hPower, const zes_power_sustained_limit_t *pSustained, const zes_power_burst_limit_t *pBurst, const zes_power_peak_limit_t *pPeak)¶ Set power limits.
- Parameters
hPower
: [in] Handle for the component.pSustained
: [in][optional] The sustained power limit. If this is null, no changes will be made to the sustained power limits.pBurst
: [in][optional] The burst power limit. If this is null, no changes will be made to the burst power limits.pPeak
: [in][optional] The peak power limit. If this is null, no changes will be made to the peak power limits.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPower
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
The device is in use, meaning that the GPU is under Over clocking, applying power limits under overclocking is not supported.
zesPowerGetEnergyThreshold¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerGetEnergyThreshold
(zes_pwr_handle_t hPower, zes_energy_threshold_t *pThreshold)¶ Get energy threshold.
- Parameters
hPower
: [in] Handle for the component.pThreshold
: [in,out] Returns information about the energy threshold setting - enabled/energy threshold/process ID.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPower
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pThreshold
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Energy threshold not supported on this power domain (check zes_power_properties_t.isEnergyThresholdSupported).
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to request this feature.
zesPowerSetEnergyThreshold¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPowerSetEnergyThreshold
(zes_pwr_handle_t hPower, double threshold)¶ Set energy threshold.
- Parameters
hPower
: [in] Handle for the component.threshold
: [in] The energy threshold to be set in joules.
An event ZES_EVENT_TYPE_FLAG_ENERGY_THRESHOLD_CROSSED will be generated when the delta energy consumed starting from this call exceeds the specified threshold. Use the function zesDeviceEventRegister() to start receiving the event.
Only one running process can control the energy threshold at a given time. If another process attempts to change the energy threshold, the error ZE_RESULT_ERROR_NOT_AVAILABLE will be returned. The function zesPowerGetEnergyThreshold() to determine the process ID currently controlling this setting.
Calling this function will remove any pending energy thresholds and start counting from the time of this call.
Once the energy threshold has been reached and the event generated, the threshold is automatically removed. It is up to the application to request a new threshold.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPower
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Energy threshold not supported on this power domain (check zes_power_properties_t.isEnergyThresholdSupported).
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to request this feature.
Another running process has set the energy threshold.
Power Structures¶
zes_power_properties_t¶
-
struct
zes_power_properties_t
¶ Properties related to device power settings.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if this resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Software can change the power limits of this domain assuming the user has permissions.
-
ze_bool_t
isEnergyThresholdSupported
¶ [out] Indicates if this power domain supports the energy threshold event (ZES_EVENT_TYPE_FLAG_ENERGY_THRESHOLD_CROSSED).
-
int32_t
defaultLimit
¶ [out] The factory default TDP power limit of the part in milliwatts. A value of -1 means that this is not known.
-
int32_t
minLimit
¶ [out] The minimum power limit in milliwatts that can be requested.
-
int32_t
maxLimit
¶ [out] The maximum power limit in milliwatts that can be requested.
-
zes_structure_type_t
zes_power_energy_counter_t¶
-
struct
zes_power_energy_counter_t
¶ Energy counter snapshot.
Average power is calculated by taking two snapshots (s1, s2) and using the equation: PowerWatts = (s2.energy - s1.energy) / (s2.timestamp - s1.timestamp)
Public Members
-
uint64_t
energy
¶ [out] The monotonic energy counter in microjoules.
-
uint64_t
timestamp
¶ [out] Microsecond timestamp when energy was captured. This timestamp should only be used to calculate delta time between snapshots of this structure. Never take the delta of this timestamp with the timestamp from a different structure since they are not guaranteed to have the same base. The absolute value of the timestamp is only valid during within the application and may be different on the next execution.
zes_power_sustained_limit_t¶
-
struct
zes_power_sustained_limit_t
¶ Sustained power limits.
The power controller (Punit) will throttle the operating frequency if the power averaged over a window (typically seconds) exceeds this limit.
zes_power_burst_limit_t¶
-
struct
zes_power_burst_limit_t
¶ Burst power limit.
The power controller (Punit) will throttle the operating frequency of the device if the power averaged over a few milliseconds exceeds a limit known as PL2. Typically PL2 > PL1 so that it permits the frequency to burst higher for short periods than would be otherwise permitted by PL1.
zes_power_peak_limit_t¶
-
struct
zes_power_peak_limit_t
¶ Peak power limit.
The power controller (Punit) will reactively/proactively throttle the operating frequency of the device when the instantaneous/100usec power exceeds this limit. The limit is known as PL4 or Psys. It expresses the maximum power that can be drawn from the power supply.
If this power limit is removed or set too high, the power supply will generate an interrupt when it detects an overcurrent condition and the power controller will throttle the device frequencies down to min. It is thus better to tune the PL4 value in order to avoid such excursions.
Public Members
-
int32_t
powerAC
¶ [in,out] power limit in milliwatts for the AC power source.
-
int32_t
powerDC
¶ [in,out] power limit in milliwatts for the DC power source. On input, this is ignored if the product does not have a battery. On output, this will be -1 if the product does not have a battery.
zes_energy_threshold_t¶
-
struct
zes_energy_threshold_t
¶ Energy threshold.
.
Public Members
-
ze_bool_t
enable
¶ [in,out] Indicates if the energy threshold is enabled.
-
double
threshold
¶ [in,out] The energy threshold in Joules. Will be 0.0 if no threshold has been set.
-
uint32_t
processId
¶ [in,out] The host process ID that set the energy threshold. Will be 0xFFFFFFFF if no threshold has been set.
Psu¶
Functions
Enumerations
Structures
Psu Functions¶
zesDeviceEnumPsus¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumPsus
(zes_device_handle_t hDevice, uint32_t *pCount, zes_psu_handle_t *phPsu)¶ Get handle of power supplies.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phPsu
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesPsuGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPsuGetProperties
(zes_psu_handle_t hPsu, zes_psu_properties_t *pProperties)¶ Get power supply properties.
- Parameters
hPsu
: [in] Handle for the component.pProperties
: [in,out] Will contain the properties of the power supply.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPsu
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesPsuGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesPsuGetState
(zes_psu_handle_t hPsu, zes_psu_state_t *pState)¶ Get current power supply state.
- Parameters
hPsu
: [in] Handle for the component.pState
: [in,out] Will contain the current state of the power supply.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hPsu
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
Psu Enums¶
zes_psu_voltage_status_t¶
-
enum
zes_psu_voltage_status_t
¶ PSU voltage status.
Values:
-
ZES_PSU_VOLTAGE_STATUS_UNKNOWN
= 0¶ The status of the power supply voltage controllers cannot be determined
-
ZES_PSU_VOLTAGE_STATUS_NORMAL
= 1¶ No unusual voltages have been detected.
-
ZES_PSU_VOLTAGE_STATUS_OVER
= 2¶ Over-voltage has occurred.
-
ZES_PSU_VOLTAGE_STATUS_UNDER
= 3¶ Under-voltage has occurred.
-
ZES_PSU_VOLTAGE_STATUS_FORCE_UINT32
= 0x7fffffff¶
-
Psu Structures¶
zes_psu_properties_t¶
-
struct
zes_psu_properties_t
¶ Static properties of the power supply.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
haveFan
¶ [out] True if the power supply has a fan
-
int32_t
ampLimit
¶ [out] The maximum electrical current in milliamperes that can be drawn. A value of -1 indicates that this property cannot be determined.
-
zes_structure_type_t
zes_psu_state_t¶
-
struct
zes_psu_state_t
¶ Dynamic state of the power supply.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
zes_psu_voltage_status_t
voltStatus
¶ [out] The current PSU voltage status
-
ze_bool_t
fanFailed
¶ [out] Indicates if the fan has failed
-
int32_t
temperature
¶ [out] Read the current heatsink temperature in degrees Celsius. A value of -1 indicates that this property cannot be determined.
-
int32_t
current
¶ [out] The amps being drawn in milliamperes. A value of -1 indicates that this property cannot be determined.
-
zes_structure_type_t
Ras¶
Functions
Enumerations
Structures
Ras Functions¶
zesDeviceEnumRasErrorSets¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumRasErrorSets
(zes_device_handle_t hDevice, uint32_t *pCount, zes_ras_handle_t *phRas)¶ Get handle of all RAS error sets on a device.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phRas
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
A RAS error set is a collection of RAS error counters of a given type (correctable/uncorrectable) from hardware blocks contained within a sub-device or within the device.
A device without sub-devices will typically return two handles, one for correctable errors sets and one for uncorrectable error sets.
A device with sub-devices will return RAS error sets for each sub-device and possibly RAS error sets for hardware blocks outside the sub-devices.
If the function completes successfully but pCount is set to 0, RAS features are not available/enabled on this device.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesRasGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesRasGetProperties
(zes_ras_handle_t hRas, zes_ras_properties_t *pProperties)¶ Get RAS properties of a given RAS error set - this enables discovery of the type of RAS error set (correctable/uncorrectable) and if located on a sub-device.
- Parameters
hRas
: [in] Handle for the component.pProperties
: [in,out] Structure describing RAS properties
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hRas
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesRasGetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesRasGetConfig
(zes_ras_handle_t hRas, zes_ras_config_t *pConfig)¶ Get RAS error thresholds that control when RAS events are generated.
- Parameters
hRas
: [in] Handle for the component.pConfig
: [in,out] Will be populed with the current RAS configuration - thresholds used to trigger events
The driver maintains counters for all RAS error sets and error categories. Events are generated when errors occur. The configuration enables setting thresholds to limit when events are sent.
When a particular RAS correctable error counter exceeds the configured threshold, the event ZES_EVENT_TYPE_FLAG_RAS_CORRECTABLE_ERRORS will be triggered.
When a particular RAS uncorrectable error counter exceeds the configured threshold, the event ZES_EVENT_TYPE_FLAG_RAS_UNCORRECTABLE_ERRORS will be triggered.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hRas
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
zesRasSetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesRasSetConfig
(zes_ras_handle_t hRas, const zes_ras_config_t *pConfig)¶ Set RAS error thresholds that control when RAS events are generated.
- Parameters
hRas
: [in] Handle for the component.pConfig
: [in] Change the RAS configuration - thresholds used to trigger events
The driver maintains counters for all RAS error sets and error categories. Events are generated when errors occur. The configuration enables setting thresholds to limit when events are sent.
When a particular RAS correctable error counter exceeds the specified threshold, the event ZES_EVENT_TYPE_FLAG_RAS_CORRECTABLE_ERRORS will be generated.
When a particular RAS uncorrectable error counter exceeds the specified threshold, the event ZES_EVENT_TYPE_FLAG_RAS_UNCORRECTABLE_ERRORS will be generated.
Call zesRasGetState() and set the clear flag to true to restart event generation once counters have exceeded thresholds.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hRas
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
Another running process is controlling these settings.
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
Don’t have permissions to set thresholds.
zesRasGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesRasGetState
(zes_ras_handle_t hRas, ze_bool_t clear, zes_ras_state_t *pState)¶ Get the current value of RAS error counters for a particular error set.
- Parameters
hRas
: [in] Handle for the component.clear
: [in] Set to 1 to clear the counters of this typepState
: [in,out] Breakdown of where errors have occurred
Clearing errors will affect other threads/applications - the counter values will start from zero.
Clearing errors requires write permissions.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hRas
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pState
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
Don’t have permissions to clear error counters.
Ras Enums¶
zes_ras_error_type_t¶
zes_ras_error_cat_t¶
-
enum
zes_ras_error_cat_t
¶ RAS error categories.
Values:
-
ZES_RAS_ERROR_CAT_RESET
= 0¶ The number of accelerator engine resets attempted by the driver.
-
ZES_RAS_ERROR_CAT_PROGRAMMING_ERRORS
= 1¶ The number of hardware exceptions generated by the way workloads have programmed the hardware
-
ZES_RAS_ERROR_CAT_DRIVER_ERRORS
= 2¶ The number of low level driver communication errors have occurred.
-
ZES_RAS_ERROR_CAT_COMPUTE_ERRORS
= 3¶ The number of errors that have occurred in the compute accelerator hardware
-
ZES_RAS_ERROR_CAT_NON_COMPUTE_ERRORS
= 4¶ The number of errors that have occurred in the fixed-function accelerator hardware
-
ZES_RAS_ERROR_CAT_CACHE_ERRORS
= 5¶ The number of errors that have occurred in caches (L1/L3/register file/shared local memory/sampler)
-
ZES_RAS_ERROR_CAT_DISPLAY_ERRORS
= 6¶ The number of errors that have occurred in the display.
-
ZES_RAS_ERROR_CAT_FORCE_UINT32
= 0x7fffffff¶
-
Ras Structures¶
zes_ras_properties_t¶
-
struct
zes_ras_properties_t
¶ RAS properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_ras_error_type_t
type
¶ [out] The type of RAS error
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_structure_type_t
zes_ras_state_t¶
-
struct
zes_ras_state_t
¶ RAS error details.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
uint64_t
category
[ZES_MAX_RAS_ERROR_CATEGORY_COUNT
]¶ [in][out] Breakdown of error by category
-
zes_structure_type_t
zes_ras_config_t¶
-
struct
zes_ras_config_t
¶ RAS error configuration - thresholds used for triggering RAS events (ZES_EVENT_TYPE_FLAG_RAS_CORRECTABLE_ERRORS, ZES_EVENT_TYPE_FLAG_RAS_UNCORRECTABLE_ERRORS)
The driver maintains a total counter which is updated every time a hardware block covered by the corresponding RAS error set notifies that an error has occurred. When this total count goes above the totalThreshold specified below, a RAS event is triggered.
The driver also maintains a counter for each category of RAS error (see zes_ras_state_t for a breakdown). Each time a hardware block of that category notifies that an error has occurred, that corresponding category counter is updated. When it goes above the threshold specified in detailedThresholds, a RAS event is triggered.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
uint64_t
totalThreshold
¶ [in,out] If the total RAS errors exceeds this threshold, the event will be triggered. A value of 0ULL disables triggering the event based on the total counter.
-
zes_ras_state_t
detailedThresholds
¶ [in,out] If the RAS errors for each category exceed the threshold for that category, the event will be triggered. A value of 0ULL will disable an event being triggered for that category.
Scheduler¶
Functions
Enumerations
Structures
Scheduler Functions¶
zesDeviceEnumSchedulers¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumSchedulers
(zes_device_handle_t hDevice, uint32_t *pCount, zes_sched_handle_t *phScheduler)¶ Returns handles to scheduler components.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phScheduler
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
Each scheduler component manages the distribution of work across one or more accelerator engines.
If an application wishes to change the scheduler behavior for all accelerator engines of a specific type (e.g. compute), it should select all the handles where the structure member zes_sched_properties_t.engines contains that type.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesSchedulerGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerGetProperties
(zes_sched_handle_t hScheduler, zes_sched_properties_t *pProperties)¶ Get properties related to a scheduler component.
- Parameters
hScheduler
: [in] Handle for the component.pProperties
: [in,out] Structure that will contain property data.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesSchedulerGetCurrentMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerGetCurrentMode
(zes_sched_handle_t hScheduler, zes_sched_mode_t *pMode)¶ Get current scheduling mode in effect on a scheduler component.
- Parameters
hScheduler
: [in] Sysman handle for the component.pMode
: [in,out] Will contain the current scheduler mode.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pMode
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
zesSchedulerGetTimeoutModeProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerGetTimeoutModeProperties
(zes_sched_handle_t hScheduler, ze_bool_t getDefaults, zes_sched_timeout_properties_t *pConfig)¶ Get scheduler config for mode ZES_SCHED_MODE_TIMEOUT.
- Parameters
hScheduler
: [in] Sysman handle for the component.getDefaults
: [in] If TRUE, the driver will return the system default properties for this mode, otherwise it will return the current properties.pConfig
: [in,out] Will contain the current parameters for this mode.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
zesSchedulerGetTimesliceModeProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerGetTimesliceModeProperties
(zes_sched_handle_t hScheduler, ze_bool_t getDefaults, zes_sched_timeslice_properties_t *pConfig)¶ Get scheduler config for mode ZES_SCHED_MODE_TIMESLICE.
- Parameters
hScheduler
: [in] Sysman handle for the component.getDefaults
: [in] If TRUE, the driver will return the system default properties for this mode, otherwise it will return the current properties.pConfig
: [in,out] Will contain the current parameters for this mode.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
zesSchedulerSetTimeoutMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerSetTimeoutMode
(zes_sched_handle_t hScheduler, zes_sched_timeout_properties_t *pProperties, ze_bool_t *pNeedReload)¶ Change scheduler mode to ZES_SCHED_MODE_TIMEOUT or update scheduler mode parameters if already running in this mode.
- Parameters
hScheduler
: [in] Sysman handle for the component.pProperties
: [in] The properties to use when configurating this mode.pNeedReload
: [in,out] Will be set to TRUE if a device driver reload is needed to apply the new scheduler mode.
This mode is optimized for multiple applications or contexts submitting work to the hardware. When higher priority work arrives, the scheduler attempts to pause the current executing work within some timeout interval, then submits the other work.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
nullptr == pNeedReload
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make this modification.
zesSchedulerSetTimesliceMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerSetTimesliceMode
(zes_sched_handle_t hScheduler, zes_sched_timeslice_properties_t *pProperties, ze_bool_t *pNeedReload)¶ Change scheduler mode to ZES_SCHED_MODE_TIMESLICE or update scheduler mode parameters if already running in this mode.
- Parameters
hScheduler
: [in] Sysman handle for the component.pProperties
: [in] The properties to use when configurating this mode.pNeedReload
: [in,out] Will be set to TRUE if a device driver reload is needed to apply the new scheduler mode.
This mode is optimized to provide fair sharing of hardware execution time between multiple contexts submitting work to the hardware concurrently.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
nullptr == pNeedReload
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make this modification.
zesSchedulerSetExclusiveMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerSetExclusiveMode
(zes_sched_handle_t hScheduler, ze_bool_t *pNeedReload)¶ Change scheduler mode to ZES_SCHED_MODE_EXCLUSIVE.
- Parameters
hScheduler
: [in] Sysman handle for the component.pNeedReload
: [in,out] Will be set to TRUE if a device driver reload is needed to apply the new scheduler mode.
This mode is optimized for single application/context use-cases. It permits a context to run indefinitely on the hardware without being preempted or terminated. All pending work for other contexts must wait until the running context completes with no further submitted work.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pNeedReload
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make this modification.
zesSchedulerSetComputeUnitDebugMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesSchedulerSetComputeUnitDebugMode
(zes_sched_handle_t hScheduler, ze_bool_t *pNeedReload)¶ Change scheduler mode to ZES_SCHED_MODE_COMPUTE_UNIT_DEBUG.
- Parameters
hScheduler
: [in] Sysman handle for the component.pNeedReload
: [in,out] Will be set to TRUE if a device driver reload is needed to apply the new scheduler mode.
This is a special mode that must ben enabled when debugging an application that uses this device e.g. using the Level0 Debug API.
It ensures that only one command queue can execute work on the hardware at a given time. Work is permitted to run as long as needed without enforcing any scheduler fairness policies.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hScheduler
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pNeedReload
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
This scheduler component does not support scheduler modes.
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make this modification.
Scheduler Enums¶
zes_sched_mode_t¶
-
enum
zes_sched_mode_t
¶ Scheduler mode.
Values:
-
ZES_SCHED_MODE_TIMEOUT
= 0¶ Multiple applications or contexts are submitting work to the hardware. When higher priority work arrives, the scheduler attempts to pause the current executing work within some timeout interval, then submits the other work.
-
ZES_SCHED_MODE_TIMESLICE
= 1¶ The scheduler attempts to fairly timeslice hardware execution time between multiple contexts submitting work to the hardware concurrently.
-
ZES_SCHED_MODE_EXCLUSIVE
= 2¶ Any application or context can run indefinitely on the hardware without being preempted or terminated. All pending work for other contexts must wait until the running context completes with no further submitted work.
-
ZES_SCHED_MODE_COMPUTE_UNIT_DEBUG
= 3¶ This is a special mode that must ben enabled when debugging an application that uses this device e.g. using the Level0 Debug API. It has the effect of disabling any timeouts on workload execution time and will change workload scheduling to ensure debug accuracy.
-
ZES_SCHED_MODE_FORCE_UINT32
= 0x7fffffff¶
-
Scheduler Structures¶
zes_sched_properties_t¶
-
struct
zes_sched_properties_t
¶ Properties related to scheduler component.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
ze_bool_t
onSubdevice
¶ [out] True if this resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
ze_bool_t
canControl
¶ [out] Software can change the scheduler component configuration assuming the user has permissions.
-
zes_engine_type_flags_t
engines
¶ [out] Bitfield of accelerator engine types that are managed by this scheduler component. Note that there can be more than one scheduler component for the same type of accelerator engine.
-
uint32_t
supportedModes
¶ [out] Bitfield of scheduler modes that can be configured for this scheduler component (bitfield of 1<<zes_sched_mode_t).
-
zes_structure_type_t
zes_sched_timeout_properties_t¶
-
struct
zes_sched_timeout_properties_t
¶ Configuration for timeout scheduler mode (ZES_SCHED_MODE_TIMEOUT)
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
uint64_t
watchdogTimeout
¶ [in,out] The maximum time in microseconds that the scheduler will wait for a batch of work submitted to a hardware engine to complete or to be preempted so as to run another context. If this time is exceeded, the hardware engine is reset and the context terminated. If set to ZES_SCHED_WATCHDOG_DISABLE, a running workload can run as long as it wants without being terminated, but preemption attempts to run other contexts are permitted but not enforced.
-
zes_structure_type_t
zes_sched_timeslice_properties_t¶
-
struct
zes_sched_timeslice_properties_t
¶ Configuration for timeslice scheduler mode (ZES_SCHED_MODE_TIMESLICE)
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
uint64_t
interval
¶ [in,out] The average interval in microseconds that a submission for a context will run on a hardware engine before being preempted out to run a pending submission for another context.
-
uint64_t
yieldTimeout
¶ [in,out] The maximum time in microseconds that the scheduler will wait to preempt a workload running on an engine before deciding to reset the hardware engine and terminating the associated context.
-
zes_structure_type_t
Standby¶
Functions
Enumerations
Structures
Standby Functions¶
zesDeviceEnumStandbyDomains¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumStandbyDomains
(zes_device_handle_t hDevice, uint32_t *pCount, zes_standby_handle_t *phStandby)¶ Get handle of standby controls.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phStandby
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesStandbyGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesStandbyGetProperties
(zes_standby_handle_t hStandby, zes_standby_properties_t *pProperties)¶ Get standby hardware component properties.
- Parameters
hStandby
: [in] Handle for the component.pProperties
: [in,out] Will contain the standby hardware properties.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hStandby
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesStandbyGetMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesStandbyGetMode
(zes_standby_handle_t hStandby, zes_standby_promo_mode_t *pMode)¶ Get the current standby promotion mode.
- Parameters
hStandby
: [in] Handle for the component.pMode
: [in,out] Will contain the current standby mode.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hStandby
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pMode
zesStandbySetMode¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesStandbySetMode
(zes_standby_handle_t hStandby, zes_standby_promo_mode_t mode)¶ Set standby promotion mode.
- Parameters
hStandby
: [in] Handle for the component.mode
: [in] New standby mode.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hStandby
ZE_RESULT_ERROR_INVALID_ENUMERATION
ZES_STANDBY_PROMO_MODE_NEVER < mode
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to make these modifications.
Standby Enums¶
zes_standby_type_t¶
zes_standby_promo_mode_t¶
-
enum
zes_standby_promo_mode_t
¶ Standby promotion modes.
Values:
-
ZES_STANDBY_PROMO_MODE_DEFAULT
= 0¶ Best compromise between performance and energy savings.
-
ZES_STANDBY_PROMO_MODE_NEVER
= 1¶ The device/component will never shutdown. This can improve performance but uses more energy.
-
ZES_STANDBY_PROMO_MODE_FORCE_UINT32
= 0x7fffffff¶
-
Standby Structures¶
zes_standby_properties_t¶
-
struct
zes_standby_properties_t
¶ Standby hardware component properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_standby_type_t
type
¶ [out] Which standby hardware component this controls
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
zes_structure_type_t
Temperature¶
Functions
Enumerations
Structures
Temperature Functions¶
zesDeviceEnumTemperatureSensors¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesDeviceEnumTemperatureSensors
(zes_device_handle_t hDevice, uint32_t *pCount, zes_temp_handle_t *phTemperature)¶ Get handle of temperature sensors.
- Parameters
hDevice
: [in] Sysman handle of the device.pCount
: [in,out] pointer to the number of components of this type. if count is zero, then the driver shall update the value with the total number of components of this type that are available. if count is greater than the number of components of this type that are available, then the driver shall update the value with the correct number of components.phTemperature
: [in,out][optional][range(0, *pCount)] array of handle of components of this type. if count is less than the number of components of this type that are available, then the driver shall only retrieve that number of component handles.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hDevice
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pCount
zesTemperatureGetProperties¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesTemperatureGetProperties
(zes_temp_handle_t hTemperature, zes_temp_properties_t *pProperties)¶ Get temperature sensor properties.
- Parameters
hTemperature
: [in] Handle for the component.pProperties
: [in,out] Will contain the temperature sensor properties.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hTemperature
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pProperties
zesTemperatureGetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesTemperatureGetConfig
(zes_temp_handle_t hTemperature, zes_temp_config_t *pConfig)¶ Get temperature configuration for this sensor - which events are triggered and the trigger conditions.
- Parameters
hTemperature
: [in] Handle for the component.pConfig
: [in,out] Returns current configuration.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hTemperature
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Temperature thresholds are not supported on this temperature sensor. Generally this is only supported for temperature sensor ZES_TEMP_SENSORS_GLOBAL
One or both of the thresholds is not supported - check zes_temp_properties_t.isThreshold1Supported and zes_temp_properties_t.isThreshold2Supported
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to request this feature.
zesTemperatureSetConfig¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesTemperatureSetConfig
(zes_temp_handle_t hTemperature, const zes_temp_config_t *pConfig)¶ Set temperature configuration for this sensor - indicates which events are triggered and the trigger conditions.
- Parameters
hTemperature
: [in] Handle for the component.pConfig
: [in] New configuration.
Events ZES_EVENT_TYPE_FLAG_TEMP_CRITICAL will be triggered when temperature reaches the critical range. Use the function zesDeviceEventRegister() to start receiving this event.
Events ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD1 and ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD2 will be generated when temperature cross the thresholds set using this function. Use the function zesDeviceEventRegister() to start receiving these events.
Only one running process can set the temperature configuration at a time. If another process attempts to change the configuration, the error ZE_RESULT_ERROR_NOT_AVAILABLE will be returned. The function zesTemperatureGetConfig() will return the process ID currently controlling these settings.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hTemperature
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pConfig
ZE_RESULT_ERROR_UNSUPPORTED_FEATURE
Temperature thresholds are not supported on this temperature sensor. Generally they are only supported for temperature sensor ZES_TEMP_SENSORS_GLOBAL
Enabling the critical temperature event is not supported - check zes_temp_properties_t.isCriticalTempSupported
One or both of the thresholds is not supported - check zes_temp_properties_t.isThreshold1Supported and zes_temp_properties_t.isThreshold2Supported
ZE_RESULT_ERROR_INSUFFICIENT_PERMISSIONS
User does not have permissions to request this feature.
Another running process is controlling these settings.
ZE_RESULT_ERROR_INVALID_ARGUMENT
One or both the thresholds is above TjMax (see zesFrequencyOcGetTjMax()). Temperature thresholds must be below this value.
zesTemperatureGetState¶
-
ZE_APIEXPORT ze_result_t ZE_APICALL
zesTemperatureGetState
(zes_temp_handle_t hTemperature, double *pTemperature)¶ Get the temperature from a specified sensor.
- Parameters
hTemperature
: [in] Handle for the component.pTemperature
: [in,out] Will contain the temperature read from the specified sensor in degrees Celsius.
The application may call this function from simultaneous threads.
The implementation of this function should be lock-free.
- Return
ZE_RESULT_ERROR_INVALID_NULL_HANDLE
nullptr == hTemperature
ZE_RESULT_ERROR_INVALID_NULL_POINTER
nullptr == pTemperature
Temperature Enums¶
zes_temp_sensors_t¶
-
enum
zes_temp_sensors_t
¶ Temperature sensors.
Values:
-
ZES_TEMP_SENSORS_GLOBAL
= 0¶ The maximum temperature across all device sensors.
-
ZES_TEMP_SENSORS_GPU
= 1¶ The maximum temperature across all sensors in the GPU.
-
ZES_TEMP_SENSORS_MEMORY
= 2¶ The maximum temperature across all sensors in the local memory.
-
ZES_TEMP_SENSORS_GLOBAL_MIN
= 3¶ The minimum temperature across all device sensors.
-
ZES_TEMP_SENSORS_GPU_MIN
= 4¶ The minimum temperature across all sensors in the GPU.
-
ZES_TEMP_SENSORS_MEMORY_MIN
= 5¶ The minimum temperature across all sensors in the local device memory.
-
ZES_TEMP_SENSORS_FORCE_UINT32
= 0x7fffffff¶
-
Temperature Structures¶
zes_temp_properties_t¶
-
struct
zes_temp_properties_t
¶ Temperature sensor properties.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
void *
pNext
¶ [in,out][optional] pointer to extension-specific structure
-
zes_temp_sensors_t
type
¶ [out] Which part of the device the temperature sensor measures
-
ze_bool_t
onSubdevice
¶ [out] True if the resource is located on a sub-device; false means that the resource is on the device of the calling Sysman handle
-
uint32_t
subdeviceId
¶ [out] If onSubdevice is true, this gives the ID of the sub-device
-
double
maxTemperature
¶ [out] Will contain the maximum temperature for the specific device in degrees Celsius.
-
ze_bool_t
isCriticalTempSupported
¶ [out] Indicates if the critical temperature event ZES_EVENT_TYPE_FLAG_TEMP_CRITICAL is supported
-
ze_bool_t
isThreshold1Supported
¶ [out] Indicates if the temperature threshold 1 event ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD1 is supported
-
ze_bool_t
isThreshold2Supported
¶ [out] Indicates if the temperature threshold 2 event ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD2 is supported
-
zes_structure_type_t
zes_temp_threshold_t¶
-
struct
zes_temp_threshold_t
¶ Temperature sensor threshold.
Public Members
-
ze_bool_t
enableLowToHigh
¶ [in,out] Trigger an event when the temperature crosses from below the threshold to above.
-
ze_bool_t
enableHighToLow
¶ [in,out] Trigger an event when the temperature crosses from above the threshold to below.
-
double
threshold
¶ [in,out] The threshold in degrees Celsius.
-
ze_bool_t
zes_temp_config_t¶
-
struct
zes_temp_config_t
¶ Temperature configuration - which events should be triggered and the trigger conditions.
Public Members
-
zes_structure_type_t
stype
¶ [in] type of this structure
-
const void *
pNext
¶ [in][optional] pointer to extension-specific structure
-
ze_bool_t
enableCritical
¶ [in,out] Indicates if event ZES_EVENT_TYPE_FLAG_TEMP_CRITICAL should be triggered by the driver.
-
zes_temp_threshold_t
threshold1
¶ [in,out] Configuration controlling if and when event ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD1 should be triggered by the driver.
-
zes_temp_threshold_t
threshold2
¶ [in,out] Configuration controlling if and when event ZES_EVENT_TYPE_FLAG_TEMP_THRESHOLD2 should be triggered by the driver.
-
zes_structure_type_t