Device Management#
Functions | |
hipError_t | hipDeviceSynchronize (void) |
Waits on all active streams on current device. | |
hipError_t | hipDeviceReset (void) |
The state of current device is discarded and updated to a fresh state. | |
hipError_t | hipSetDevice (int deviceId) |
Set default device to be used for subsequent hip API calls from this thread. | |
hipError_t | hipGetDevice (int *deviceId) |
Return the default device id for the calling host thread. | |
hipError_t | hipGetDeviceCount (int *count) |
Return number of compute-capable devices. | |
hipError_t | hipDeviceGetAttribute (int *pi, hipDeviceAttribute_t attr, int deviceId) |
Query for a specific device attribute. | |
hipError_t | hipDeviceGetDefaultMemPool (hipMemPool_t *mem_pool, int device) |
Returns the default memory pool of the specified device. | |
hipError_t | hipDeviceSetMemPool (int device, hipMemPool_t mem_pool) |
Sets the current memory pool of a device. | |
hipError_t | hipDeviceGetMemPool (hipMemPool_t *mem_pool, int device) |
Gets the current memory pool for the specified device. | |
hipError_t | hipGetDeviceProperties (hipDeviceProp_t *prop, int deviceId) |
Returns device properties. | |
hipError_t | hipDeviceSetCacheConfig (hipFuncCache_t cacheConfig) |
Set L1/Shared cache partition. | |
hipError_t | hipDeviceGetCacheConfig (hipFuncCache_t *cacheConfig) |
Get Cache configuration for a specific Device. | |
hipError_t | hipDeviceGetLimit (size_t *pValue, enum hipLimit_t limit) |
Gets resource limits of current device. | |
hipError_t | hipDeviceSetLimit (enum hipLimit_t limit, size_t value) |
Sets resource limits of current device. | |
hipError_t | hipDeviceGetSharedMemConfig (hipSharedMemConfig *pConfig) |
Returns bank width of shared memory for current device. | |
hipError_t | hipGetDeviceFlags (unsigned int *flags) |
Gets the flags set for current device. | |
hipError_t | hipDeviceSetSharedMemConfig (hipSharedMemConfig config) |
The bank width of shared memory on current device is set. | |
hipError_t | hipSetDeviceFlags (unsigned flags) |
The current device behavior is changed according the flags passed. | |
hipError_t | hipChooseDevice (int *device, const hipDeviceProp_t *prop) |
Device which matches hipDeviceProp_t is returned. | |
hipError_t | hipExtGetLinkTypeAndHopCount (int device1, int device2, uint32_t *linktype, uint32_t *hopcount) |
Returns the link type and hop count between two devices. | |
hipError_t | hipIpcGetMemHandle (hipIpcMemHandle_t *handle, void *devPtr) |
Gets an interprocess memory handle for an existing device memory allocation. | |
hipError_t | hipIpcOpenMemHandle (void **devPtr, hipIpcMemHandle_t handle, unsigned int flags) |
Opens an interprocess memory handle exported from another process and returns a device pointer usable in the local process. | |
hipError_t | hipIpcCloseMemHandle (void *devPtr) |
Close memory mapped with hipIpcOpenMemHandle. | |
hipError_t | hipIpcGetEventHandle (hipIpcEventHandle_t *handle, hipEvent_t event) |
Gets an opaque interprocess handle for an event. | |
hipError_t | hipIpcOpenEventHandle (hipEvent_t *event, hipIpcEventHandle_t handle) |
Opens an interprocess event handles. | |
Detailed Description
This section describes the device management functions of HIP runtime API.
Function Documentation
◆ hipChooseDevice()
hipError_t hipChooseDevice | ( | int * | device, |
const hipDeviceProp_t * | prop | ||
) |
Device which matches hipDeviceProp_t is returned.
- Parameters
-
[out] device Pointer of the device [in] prop Pointer of the properties
- Returns
- hipSuccess, hipErrorInvalidValue
◆ hipDeviceGetAttribute()
hipError_t hipDeviceGetAttribute | ( | int * | pi, |
hipDeviceAttribute_t | attr, | ||
int | deviceId | ||
) |
Query for a specific device attribute.
- Parameters
-
[out] pi pointer to value to return [in] attr attribute to query [in] deviceId which device to query for information
◆ hipDeviceGetCacheConfig()
hipError_t hipDeviceGetCacheConfig | ( | hipFuncCache_t * | cacheConfig | ) |
Get Cache configuration for a specific Device.
- Parameters
-
[out] cacheConfig Pointer of cache configuration
- Returns
- hipSuccess, hipErrorNotInitialized Note: AMD devices do not support reconfigurable cache. This hint is ignored on these architectures.
◆ hipDeviceGetDefaultMemPool()
hipError_t hipDeviceGetDefaultMemPool | ( | hipMemPool_t * | mem_pool, |
int | device | ||
) |
Returns the default memory pool of the specified device.
- Parameters
-
[out] mem_pool Default memory pool to return [in] device Device index for query the default memory pool
- See also
- hipDeviceGetDefaultMemPool, hipMallocAsync, hipMemPoolTrimTo, hipMemPoolGetAttribute, hipDeviceSetMemPool, hipMemPoolSetAttribute, hipMemPoolSetAccess, hipMemPoolGetAccess
- Warning
- : This API is marked as beta, meaning, while this is feature complete, it is still open to changes and may have outstanding issues.
◆ hipDeviceGetLimit()
hipError_t hipDeviceGetLimit | ( | size_t * | pValue, |
enum hipLimit_t | limit | ||
) |
Gets resource limits of current device.
The function queries the size of limit value, as required by the input enum value hipLimit_t, which can be either hipLimitStackSize, or hipLimitMallocHeapSize. Any other input as default, the function will return hipErrorUnsupportedLimit.
- Parameters
-
[out] pValue Returns the size of the limit in bytes [in] limit The limit to query
◆ hipDeviceGetMemPool()
hipError_t hipDeviceGetMemPool | ( | hipMemPool_t * | mem_pool, |
int | device | ||
) |
Gets the current memory pool for the specified device.
Returns the last pool provided to hipDeviceSetMemPool
for this device or the device's default memory pool if hipDeviceSetMemPool
has never been called. By default the current mempool is the default mempool for a device, otherwise the returned pool must have been set with hipDeviceSetMemPool
.
- Parameters
-
[out] mem_pool Current memory pool on the specified device [in] device Device index to query the current memory pool
- See also
- hipDeviceGetDefaultMemPool, hipMallocAsync, hipMemPoolTrimTo, hipMemPoolGetAttribute, hipDeviceSetMemPool, hipMemPoolSetAttribute, hipMemPoolSetAccess, hipMemPoolGetAccess
- Warning
- : This API is marked as beta, meaning, while this is feature complete, it is still open to changes and may have outstanding issues.
◆ hipDeviceGetSharedMemConfig()
hipError_t hipDeviceGetSharedMemConfig | ( | hipSharedMemConfig * | pConfig | ) |
Returns bank width of shared memory for current device.
- Parameters
-
[out] pConfig The pointer of the bank width for shared memory
Note: AMD devices and some Nvidia GPUS do not support shared cache banking, and the hint is ignored on those architectures.
◆ hipDeviceReset()
hipError_t hipDeviceReset | ( | void | ) |
The state of current device is discarded and updated to a fresh state.
Calling this function deletes all streams created, memory allocated, kernels running, events created. Make sure that no other thread is using the device or streams, memory, kernels, events associated with the current device.
- Returns
- hipSuccess
- See also
- hipDeviceSynchronize
◆ hipDeviceSetCacheConfig()
hipError_t hipDeviceSetCacheConfig | ( | hipFuncCache_t | cacheConfig | ) |
Set L1/Shared cache partition.
- Parameters
-
[in] cacheConfig Cache configuration
Note: AMD devices do not support reconfigurable cache. This API is not implemented on AMD platform. If the function is called, it will return hipErrorNotSupported.
◆ hipDeviceSetLimit()
hipError_t hipDeviceSetLimit | ( | enum hipLimit_t | limit, |
size_t | value | ||
) |
Sets resource limits of current device.
As the input enum limit, hipLimitStackSize sets the limit value of the stack size on the current GPU device, per thread. The limit size can get via hipDeviceGetLimit. The size is in units of 256 dwords, up to the limit (128K - 16).
hipLimitMallocHeapSize sets the limit value of the heap used by the malloc()/free() calls. The limit size can get via hipDeviceGetLimit.
Any other input as default, the funtion will return hipErrorUnsupportedLimit.
- Parameters
-
[in] limit Enum of hipLimit_t to set [in] value The size of limit value in bytes
◆ hipDeviceSetMemPool()
hipError_t hipDeviceSetMemPool | ( | int | device, |
hipMemPool_t | mem_pool | ||
) |
Sets the current memory pool of a device.
The memory pool must be local to the specified device. hipMallocAsync
allocates from the current mempool of the provided stream's device. By default, a device's current memory pool is its default memory pool.
- Note
- Use
hipMallocFromPoolAsync
for asynchronous memory allocations from a device different than the one the stream runs on.
- Parameters
-
[in] device Device index for the update [in] mem_pool Memory pool for update as the current on the specified device
- See also
- hipDeviceGetDefaultMemPool, hipMallocAsync, hipMemPoolTrimTo, hipMemPoolGetAttribute, hipDeviceSetMemPool, hipMemPoolSetAttribute, hipMemPoolSetAccess, hipMemPoolGetAccess
- Warning
- : This API is marked as beta, meaning, while this is feature complete, it is still open to changes and may have outstanding issues.
◆ hipDeviceSetSharedMemConfig()
hipError_t hipDeviceSetSharedMemConfig | ( | hipSharedMemConfig | config | ) |
The bank width of shared memory on current device is set.
- Parameters
-
[in] config Configuration for the bank width of shared memory
Note: AMD devices and some Nvidia GPUS do not support shared cache banking, and the hint is ignored on those architectures.
◆ hipDeviceSynchronize()
hipError_t hipDeviceSynchronize | ( | void | ) |
Waits on all active streams on current device.
When this command is invoked, the host thread gets blocked until all the commands associated with streams associated with the device. HIP does not support multiple blocking modes (yet!).
- Returns
- hipSuccess
- See also
- hipSetDevice, hipDeviceReset
◆ hipExtGetLinkTypeAndHopCount()
hipError_t hipExtGetLinkTypeAndHopCount | ( | int | device1, |
int | device2, | ||
uint32_t * | linktype, | ||
uint32_t * | hopcount | ||
) |
Returns the link type and hop count between two devices.
- Parameters
-
[in] device1 Ordinal for device1 [in] device2 Ordinal for device2 [out] linktype Returns the link type (See hsa_amd_link_info_type_t) between the two devices [out] hopcount Returns the hop count between the two devices
Queries and returns the HSA link type and the hop count between the two specified devices.
- Returns
- hipSuccess, #hipInvalidDevice, hipErrorRuntimeOther
◆ hipGetDevice()
hipError_t hipGetDevice | ( | int * | deviceId | ) |
Return the default device id for the calling host thread.
- Parameters
-
[out] deviceId *device is written with the default device
HIP maintains an default device for each thread using thread-local-storage. This device is used implicitly for HIP runtime APIs called by this thread. hipGetDevice returns in * device
the default device for the calling host thread.
- See also
- hipSetDevice, hipGetDevicesizeBytes
◆ hipGetDeviceCount()
hipError_t hipGetDeviceCount | ( | int * | count | ) |
Return number of compute-capable devices.
- Parameters
-
[out] count Returns number of compute-capable devices.
- Returns
- hipSuccess, hipErrorNoDevice
Returns in *count
the number of devices that have ability to run compute commands. If there are no such devices, then hipGetDeviceCount will return hipErrorNoDevice. If 1 or more devices can be found, then hipGetDeviceCount returns hipSuccess.
◆ hipGetDeviceFlags()
hipError_t hipGetDeviceFlags | ( | unsigned int * | flags | ) |
Gets the flags set for current device.
- Parameters
-
[out] flags Pointer of the flags
◆ hipGetDeviceProperties()
hipError_t hipGetDeviceProperties | ( | hipDeviceProp_t * | prop, |
int | deviceId | ||
) |
Returns device properties.
- Parameters
-
[out] prop written with device properties [in] deviceId which device to query for information
- Returns
- hipSuccess, hipErrorInvalidDevice
- Bug:
HCC always returns 0 for maxThreadsPerMultiProcessor
HCC always returns 0 for regsPerBlock
HCC always returns 0 for l2CacheSize
Populates hipGetDeviceProperties with information for the specified device.
◆ hipIpcCloseMemHandle()
hipError_t hipIpcCloseMemHandle | ( | void * | devPtr | ) |
Close memory mapped with hipIpcOpenMemHandle.
Unmaps memory returnd by hipIpcOpenMemHandle. The original allocation in the exporting process as well as imported mappings in other processes will be unaffected.
Any resources used to enable peer access will be freed if this is the last mapping using them.
- Parameters
-
devPtr - Device pointer returned by hipIpcOpenMemHandle
- Note
- This IPC memory related feature API on Windows may behave differently from Linux.
◆ hipIpcGetEventHandle()
hipError_t hipIpcGetEventHandle | ( | hipIpcEventHandle_t * | handle, |
hipEvent_t | event | ||
) |
Gets an opaque interprocess handle for an event.
This opaque handle may be copied into other processes and opened with hipIpcOpenEventHandle. Then hipEventRecord, hipEventSynchronize, hipStreamWaitEvent and hipEventQuery may be used in either process. Operations on the imported event after the exported event has been freed with hipEventDestroy will result in undefined behavior.
- Parameters
-
[out] handle Pointer to hipIpcEventHandle to return the opaque event handle [in] event Event allocated with hipEventInterprocess and hipEventDisableTiming flags
- Note
- This IPC event related feature API is currently applicable on Linux.
◆ hipIpcGetMemHandle()
hipError_t hipIpcGetMemHandle | ( | hipIpcMemHandle_t * | handle, |
void * | devPtr | ||
) |
Gets an interprocess memory handle for an existing device memory allocation.
Takes a pointer to the base of an existing device memory allocation created with hipMalloc and exports it for use in another process. This is a lightweight operation and may be called multiple times on an allocation without adverse effects.
If a region of memory is freed with hipFree and a subsequent call to hipMalloc returns memory with the same device address, hipIpcGetMemHandle will return a unique handle for the new memory.
- Parameters
-
handle - Pointer to user allocated hipIpcMemHandle to return the handle in. devPtr - Base pointer to previously allocated device memory
- Note
- This IPC memory related feature API on Windows may behave differently from Linux.
◆ hipIpcOpenEventHandle()
hipError_t hipIpcOpenEventHandle | ( | hipEvent_t * | event, |
hipIpcEventHandle_t | handle | ||
) |
Opens an interprocess event handles.
Opens an interprocess event handle exported from another process with cudaIpcGetEventHandle. The returned hipEvent_t behaves like a locally created event with the hipEventDisableTiming flag specified. This event need be freed with hipEventDestroy. Operations on the imported event after the exported event has been freed with hipEventDestroy will result in undefined behavior. If the function is called within the same process where handle is returned by hipIpcGetEventHandle, it will return hipErrorInvalidContext.
- Parameters
-
[out] event Pointer to hipEvent_t to return the event [in] handle The opaque interprocess handle to open
- Note
- This IPC event related feature API is currently applicable on Linux.
◆ hipIpcOpenMemHandle()
hipError_t hipIpcOpenMemHandle | ( | void ** | devPtr, |
hipIpcMemHandle_t | handle, | ||
unsigned int | flags | ||
) |
Opens an interprocess memory handle exported from another process and returns a device pointer usable in the local process.
Maps memory exported from another process with hipIpcGetMemHandle into the current device address space. For contexts on different devices hipIpcOpenMemHandle can attempt to enable peer access between the devices as if the user called hipDeviceEnablePeerAccess. This behavior is controlled by the hipIpcMemLazyEnablePeerAccess flag. hipDeviceCanAccessPeer can determine if a mapping is possible.
Contexts that may open hipIpcMemHandles are restricted in the following way. hipIpcMemHandles from each device in a given process may only be opened by one context per device per other process.
Memory returned from hipIpcOpenMemHandle must be freed with hipIpcCloseMemHandle.
Calling hipFree on an exported memory region before calling hipIpcCloseMemHandle in the importing context will result in undefined behavior.
- Parameters
-
devPtr - Returned device pointer handle - hipIpcMemHandle to open flags - Flags for this operation. Must be specified as hipIpcMemLazyEnablePeerAccess
- Returns
- hipSuccess, hipErrorMapFailed, hipErrorInvalidHandle, #hipErrorTooManyPeers
- Note
- During multiple processes, using the same memory handle opened by the current context, there is no guarantee that the same device poiter will be returned in
*devPtr
. This is diffrent from CUDA. - This IPC memory related feature API on Windows may behave differently from Linux.
◆ hipSetDevice()
hipError_t hipSetDevice | ( | int | deviceId | ) |
Set default device to be used for subsequent hip API calls from this thread.
- Parameters
-
[in] deviceId Valid device in range 0...hipGetDeviceCount().
Sets device
as the default device for the calling host thread. Valid device id's are 0... (hipGetDeviceCount()-1).
Many HIP APIs implicitly use the "default device" :
- Any device memory subsequently allocated from this host thread (using hipMalloc) will be allocated on device.
- Any streams or events created from this host thread will be associated with device.
- Any kernels launched from this host thread (using hipLaunchKernel) will be executed on device (unless a specific stream is specified, in which case the device associated with that stream will be used).
This function may be called from any host thread. Multiple host threads may use the same device. This function does no synchronization with the previous or new device, and has very little runtime overhead. Applications can use hipSetDevice to quickly switch the default device before making a HIP runtime call which uses the default device.
The default device is stored in thread-local-storage for each thread. Thread-pool implementations may inherit the default device of the previous thread. A good practice is to always call hipSetDevice at the start of HIP coding sequency to establish a known standard device.
- Returns
- hipSuccess, hipErrorInvalidDevice, #hipErrorDeviceAlreadyInUse
- See also
- hipGetDevice, hipGetDeviceCount
◆ hipSetDeviceFlags()
hipError_t hipSetDeviceFlags | ( | unsigned | flags | ) |
The current device behavior is changed according the flags passed.
- Parameters
-
[in] flags Flag to set on the current device
The schedule flags impact how HIP waits for the completion of a command running on a device. hipDeviceScheduleSpin : HIP runtime will actively spin in the thread which submitted the work until the command completes. This offers the lowest latency, but will consume a CPU core and may increase power. hipDeviceScheduleYield : The HIP runtime will yield the CPU to system so that other tasks can use it. This may increase latency to detect the completion but will consume less power and is friendlier to other tasks in the system. hipDeviceScheduleBlockingSync : On ROCm platform, this is a synonym for hipDeviceScheduleYield. hipDeviceScheduleAuto : Use a hueristic to select between Spin and Yield modes. If the number of HIP contexts is greater than the number of logical processors in the system, use Spin scheduling. Else use Yield scheduling.
hipDeviceMapHost : Allow mapping host memory. On ROCM, this is always allowed and the flag is ignored. hipDeviceLmemResizeToMax :
- Warning
- ROCm silently ignores this flag.