Copyright	[2009..2023] Trevor L. McDonell
License	BSD
Safe Haskell	Safe-Inferred
Language	Haskell98

Foreign.CUDA.Analysis.Device

Description

Common device functions

Synopsis

data Compute = Compute !Int !Int
data ComputeMode
- = Default
- | Prohibited
- | ExclusiveProcess
data DeviceProperties = DeviceProperties {
- deviceName :: !String
- computeCapability :: !Compute
- totalGlobalMem :: !Int64
- totalConstMem :: !Int64
- sharedMemPerBlock :: !Int64
- regsPerBlock :: !Int
- warpSize :: !Int
- maxThreadsPerBlock :: !Int
- maxThreadsPerMultiProcessor :: !Int
- maxBlockSize :: !(Int, Int, Int)
- maxGridSize :: !(Int, Int, Int)
- maxTextureDim1D :: !Int
- maxTextureDim2D :: !(Int, Int)
- maxTextureDim3D :: !(Int, Int, Int)
- clockRate :: !Int
- multiProcessorCount :: !Int
- memPitch :: !Int64
- memBusWidth :: !Int
- memClockRate :: !Int
- textureAlignment :: !Int64
- computeMode :: !ComputeMode
- deviceOverlap :: !Bool
- concurrentKernels :: !Bool
- eccEnabled :: !Bool
- asyncEngineCount :: !Int
- cacheMemL2 :: !Int
- pciInfo :: !PCI
- tccDriverEnabled :: !Bool
- kernelExecTimeoutEnabled :: !Bool
- integrated :: !Bool
- canMapHostMemory :: !Bool
- unifiedAddressing :: !Bool
- streamPriorities :: !Bool
- globalL1Cache :: !Bool
- localL1Cache :: !Bool
- managedMemory :: !Bool
- multiGPUBoard :: !Bool
- multiGPUBoardGroupID :: !Int
- preemption :: !Bool
- singleToDoublePerfRatio :: !Int
- cooperativeLaunch :: !Bool
- cooperativeLaunchMultiDevice :: !Bool
}
data DeviceResources = DeviceResources {
- threadsPerWarp :: !Int
- coresPerMP :: !Int
- warpsPerMP :: !Int
- threadsPerMP :: !Int
- threadBlocksPerMP :: !Int
- sharedMemPerMP :: !Int
- maxSharedMemPerBlock :: !Int
- regFileSizePerMP :: !Int
- maxRegPerBlock :: !Int
- regAllocUnit :: !Int
- regAllocationStyle :: !Allocation
- maxRegPerThread :: !Int
- sharedMemAllocUnit :: !Int
- warpAllocUnit :: !Int
- warpRegAllocUnit :: !Int
- maxGridsPerDevice :: !Int
}
data Allocation
- = Warp
- | Block
data PCI = PCI {
- busID :: !Int
- deviceID :: !Int
- domainID :: !Int
}
deviceResources :: DeviceProperties -> DeviceResources
describe :: Describe a => a -> String

Documentation

data Compute Source #

GPU compute capability, major and minor revision number respectively.

Constructors

Compute !Int !Int

Instances

Instances details

Show Compute Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> Compute -> ShowS # show :: Compute -> String # showList :: [Compute] -> ShowS #
Eq Compute Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods (==) :: Compute -> Compute -> Bool # (/=) :: Compute -> Compute -> Bool #
Ord Compute Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods compare :: Compute -> Compute -> Ordering # (<) :: Compute -> Compute -> Bool # (<=) :: Compute -> Compute -> Bool # (>) :: Compute -> Compute -> Bool # (>=) :: Compute -> Compute -> Bool # max :: Compute -> Compute -> Compute # min :: Compute -> Compute -> Compute #

data ComputeMode Source #

The compute mode the device is currently in

Constructors

Default
Prohibited
ExclusiveProcess

Instances

Instances details

Enum ComputeMode Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods succ :: ComputeMode -> ComputeMode # pred :: ComputeMode -> ComputeMode # toEnum :: Int -> ComputeMode # fromEnum :: ComputeMode -> Int # enumFrom :: ComputeMode -> [ComputeMode] # enumFromThen :: ComputeMode -> ComputeMode -> [ComputeMode] # enumFromTo :: ComputeMode -> ComputeMode -> [ComputeMode] # enumFromThenTo :: ComputeMode -> ComputeMode -> ComputeMode -> [ComputeMode] #
Show ComputeMode Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> ComputeMode -> ShowS # show :: ComputeMode -> String # showList :: [ComputeMode] -> ShowS #
Describe ComputeMode Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods describe :: ComputeMode -> String Source #
Eq ComputeMode Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods (==) :: ComputeMode -> ComputeMode -> Bool # (/=) :: ComputeMode -> ComputeMode -> Bool #

data DeviceProperties Source #

Some properties of a compute device. Originally, this mirrored struct cudaDeviceProp in CUDA, but since CUDA 13 some fields have been removed from that struct, and this data type keeps them for backwards compatibility.

There are more device "properties" than those listed in this data type; use attribute to query them.

Constructors

DeviceProperties

Fields

deviceName :: !String
Identifier
computeCapability :: !Compute
Supported compute capability
totalGlobalMem :: !Int64
Available global memory on the device in bytes
totalConstMem :: !Int64
Available constant memory on the device in bytes
sharedMemPerBlock :: !Int64
Available shared memory per block in bytes
regsPerBlock :: !Int
32-bit registers per block
warpSize :: !Int
Warp size in threads (SIMD width)
maxThreadsPerBlock :: !Int
Maximum number of threads per block
maxThreadsPerMultiProcessor :: !Int
Maximum number of threads per multiprocessor
maxBlockSize :: !(Int, Int, Int)
Maximum size of each dimension of a block
maxGridSize :: !(Int, Int, Int)
Maximum size of each dimension of a grid
maxTextureDim1D :: !Int
Maximum texture dimensions
maxTextureDim2D :: !(Int, Int)
maxTextureDim3D :: !(Int, Int, Int)
clockRate :: !Int
Clock frequency in kilohertz
multiProcessorCount :: !Int
Number of multiprocessors on the device
memPitch :: !Int64
Maximum pitch in bytes allowed by memory copies
memBusWidth :: !Int
Global memory bus width in bits
memClockRate :: !Int
Peak memory clock frequency in kilohertz
textureAlignment :: !Int64
Alignment requirement for textures
computeMode :: !ComputeMode
deviceOverlap :: !Bool
Device can concurrently copy memory and execute a kernel
concurrentKernels :: !Bool
Device can possibly execute multiple kernels concurrently
eccEnabled :: !Bool
Device supports and has enabled error correction
asyncEngineCount :: !Int
Number of asynchronous engines
cacheMemL2 :: !Int
Size of the L2 cache in bytes
pciInfo :: !PCI
PCI device information for the device
tccDriverEnabled :: !Bool
Whether this is a Tesla device using the TCC driver
kernelExecTimeoutEnabled :: !Bool
Whether there is a runtime limit on kernels
integrated :: !Bool
As opposed to discrete
canMapHostMemory :: !Bool
Device can use pinned memory
unifiedAddressing :: !Bool
Device shares a unified address space with the host
streamPriorities :: !Bool
Device supports stream priorities
globalL1Cache :: !Bool
Device supports caching globals in L1 cache
localL1Cache :: !Bool
Device supports caching locals in L1 cache
managedMemory :: !Bool
Device supports allocating managed memory on this system
multiGPUBoard :: !Bool
Device is on a multi-GPU board
multiGPUBoardGroupID :: !Int
Unique identifier for a group of devices associated with the same board
preemption :: !Bool
Device supports compute pre-emption
singleToDoublePerfRatio :: !Int
Ratio of single precision performance (in floating-point operations per second) to double precision performance
cooperativeLaunch :: !Bool
Device supports launching cooperative kernels
cooperativeLaunchMultiDevice :: !Bool
Device can participate in cooperative multi-device kernels

Instances

Instances details

Show DeviceProperties Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> DeviceProperties -> ShowS # show :: DeviceProperties -> String # showList :: [DeviceProperties] -> ShowS #

data DeviceResources Source #

Constructors

DeviceResources

Fields

threadsPerWarp :: !Int
Warp size
coresPerMP :: !Int
Number of SIMD arithmetic units per multiprocessor
warpsPerMP :: !Int
Maximum number of in-flight warps per multiprocessor
threadsPerMP :: !Int
Maximum number of in-flight threads on a multiprocessor
threadBlocksPerMP :: !Int
Maximum number of thread blocks resident on a multiprocessor
sharedMemPerMP :: !Int
Total amount of shared memory per multiprocessor (bytes)
maxSharedMemPerBlock :: !Int
Maximum amount of shared memory per thread block (bytes)
regFileSizePerMP :: !Int
Total number of registers in a multiprocessor
maxRegPerBlock :: !Int
Maximum number of registers per block
regAllocUnit :: !Int
Register allocation unit size
regAllocationStyle :: !Allocation
How multiprocessor resources are divided (register allocation granularity)
maxRegPerThread :: !Int
Maximum number of registers per thread
sharedMemAllocUnit :: !Int
Shared memory allocation unit size (bytes)
warpAllocUnit :: !Int
Warp allocation granularity
warpRegAllocUnit :: !Int
Warp register allocation granularity
maxGridsPerDevice :: !Int
Maximum number of resident grids per device (concurrent kernels)

Instances

Instances details

Show DeviceResources Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> DeviceResources -> ShowS # show :: DeviceResources -> String # showList :: [DeviceResources] -> ShowS #

data Allocation Source #

Constructors

Warp
Block

Instances

Instances details

Show Allocation Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> Allocation -> ShowS # show :: Allocation -> String # showList :: [Allocation] -> ShowS #

data PCI Source #

Constructors

PCI
Fields busID :: !Int PCI bus ID of the device deviceID :: !Int PCI device ID domainID :: !Int PCI domain ID

Instances

Instances details

Show PCI Source #
Instance details Defined in Foreign.CUDA.Analysis.Device Methods showsPrec :: Int -> PCI -> ShowS # show :: PCI -> String # showList :: [PCI] -> ShowS #

deviceResources :: DeviceProperties -> DeviceResources Source #

Extract some additional hardware resource limitations for a given device.

describe :: Describe a => a -> String Source #