6.4. Coredump Attributes Control API

This section describes the coredump attribute control functions of the low-level CUDA driver application programming interface.

Enumerations

CUCoredumpGenerationFlags: Flags for controlling coredump contents.
CUcoredumpSettings: Flags for choosing a coredump attribute to get/set.

Functions

CUresult cuCoredumpDeregisterCompleteCallback(CUcoredumpCallbackHandle callback): Deregister a previously registered coredump complete callback.
CUresult cuCoredumpDeregisterStartCallback(CUcoredumpCallbackHandle callback): Deregister a previously registered coredump start callback.
CUresult cuCoredumpGetAttribute(CUcoredumpSettings attrib, void *value, size_t *size): Allows caller to fetch a coredump attribute value for the current context.
CUresult cuCoredumpGetAttributeGlobal(CUcoredumpSettings attrib, void *value, size_t *size): Allows caller to fetch a coredump attribute value for the entire application.
CUresult cuCoredumpRegisterCompleteCallback(CUcoredumpStatusCallback callback, void *userData, CUcoredumpCallbackHandle *callbackOut): Register a callback to be invoked when a GPU coredump completes.
CUresult cuCoredumpRegisterStartCallback(CUcoredumpStatusCallback callback, void *userData, CUcoredumpCallbackHandle *callbackOut): Register a callback to be invoked when a GPU coredump begins.
CUresult cuCoredumpSetAttribute(CUcoredumpSettings attrib, void *value, size_t *size): Allows caller to set a coredump attribute value for the current context.
CUresult cuCoredumpSetAttributeGlobal(CUcoredumpSettings attrib, void *value, size_t *size): Allows caller to set a coredump attribute value globally.

Typedefs

CUcoredumpCallbackHandle: Opaque handle representing a registered coredump status callback.
CUcoredumpStatusCallback: Callback function prototype for GPU coredump status notifications.

6.4.1. Enumerations

enum CUCoredumpGenerationFlags

Flags for controlling coredump contents.

Values:

enumerator CU_COREDUMP_DEFAULT_FLAGS

enumerator CU_COREDUMP_SKIP_NONRELOCATED_ELF_IMAGES

enumerator CU_COREDUMP_SKIP_GLOBAL_MEMORY

enumerator CU_COREDUMP_SKIP_SHARED_MEMORY

enumerator CU_COREDUMP_SKIP_LOCAL_MEMORY

enumerator CU_COREDUMP_SKIP_ABORT

enumerator CU_COREDUMP_SKIP_CONSTBANK_MEMORY

enumerator CU_COREDUMP_GZIP_COMPRESS

enumerator CU_COREDUMP_FAULTED_CONTEXTS_ONLY

enumerator CU_COREDUMP_SKIP_TENSOR_MEMORY

enumerator CU_COREDUMP_NO_ERRBAR_AT_EXIT

enumerator CU_COREDUMP_LOG_ONLY

enumerator CU_COREDUMP_LIGHTWEIGHT_FLAGS

enum CUcoredumpSettings

Flags for choosing a coredump attribute to get/set.

Values:

enumerator CU_COREDUMP_ENABLE_ON_EXCEPTION

enumerator CU_COREDUMP_TRIGGER_HOST

enumerator CU_COREDUMP_LIGHTWEIGHT

enumerator CU_COREDUMP_ENABLE_USER_TRIGGER

enumerator CU_COREDUMP_FILE

enumerator CU_COREDUMP_PIPE

enumerator CU_COREDUMP_GENERATION_FLAGS

enumerator CU_COREDUMP_MAX

6.4.2. Functions

CUresult cuCoredumpDeregisterCompleteCallback(CUcoredumpCallbackHandle callback)

Deregister a previously registered coredump complete callback.

This function removes a callback that was registered with cuCoredumpRegisterCompleteCallback. The callback handle becomes invalid after this call.

See also

cuCoredumpRegisterCompleteCallback

Note

It is the caller’s responsibility to deregister callbacks before they go out of scope.

Parameters: callback – - The callback handle to deregister
Returns: CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE

CUresult cuCoredumpDeregisterStartCallback(CUcoredumpCallbackHandle callback)

Deregister a previously registered coredump start callback.

This function removes a callback that was registered with cuCoredumpRegisterStartCallback. The callback handle becomes invalid after this call.

See also

cuCoredumpRegisterStartCallback

Note

It is the caller’s responsibility to deregister callbacks before they go out of scope.

Parameters: callback – - The callback handle to deregister
Returns: CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE

CUresult cuCoredumpGetAttribute(CUcoredumpSettings attrib, void *value, size_t *size)

Allows caller to fetch a coredump attribute value for the current context.

Returns in *value the requested value specified by attrib. It is up to the caller to ensure that the data type and size of *value matches the request.

If the caller calls this function with *value equal to NULL, the size of the memory region (in bytes) expected for attrib will be placed in size.

The supported attributes are:

CU_COREDUMP_ENABLE_ON_EXCEPTION: Bool where ::true means that GPU exceptions from this context will create a coredump at the location specified by CU_COREDUMP_FILE. The default value is ::false unless set to ::true globally or locally, or the CU_CTX_USER_COREDUMP_ENABLE flag was set during context creation.
CU_COREDUMP_TRIGGER_HOST: Bool where ::true means that the host CPU will also create a coredump. The default value is ::true unless set to ::false globally or or locally. This value is deprecated as of CUDA 12.5 - raise the CU_COREDUMP_SKIP_ABORT flag to disable host device abort() if needed.
CU_COREDUMP_LIGHTWEIGHT: Bool where ::true means that any resulting coredumps will not have a dump of GPU memory or non-reloc ELF images. The default value is ::false unless set to ::true globally or locally. This attribute is deprecated as of CUDA 12.5, please use CU_COREDUMP_GENERATION_FLAGS instead.
CU_COREDUMP_ENABLE_USER_TRIGGER: Bool where ::true means that a coredump can be created by writing to the system pipe specified by CU_COREDUMP_PIPE. The default value is ::false unless set to ::true globally or locally.
CU_COREDUMP_FILE: String of up to 1023 characters that defines the location where any coredumps generated by this context will be written. The default value is ::core.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA applications and ::PID is the process ID of the CUDA application.
CU_COREDUMP_PIPE: String of up to 1023 characters that defines the name of the pipe that will be monitored if user-triggered coredumps are enabled. The default value is ::corepipe.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA application and ::PID is the process ID of the CUDA application.
CU_COREDUMP_GENERATION_FLAGS: An integer with values to allow granular control the data contained in a coredump specified as a bitwise OR combination of the following values:
- CU_COREDUMP_DEFAULT_FLAGS - if set by itself, coredump generation returns to its default settings of including all memory regions that it is able to access
- CU_COREDUMP_SKIP_NONRELOCATED_ELF_IMAGES - Coredump will not include the data from CUDA source modules that are not relocated at runtime.
- CU_COREDUMP_SKIP_GLOBAL_MEMORY - Coredump will not include device-side global data that does not belong to any context.
- CU_COREDUMP_SKIP_SHARED_MEMORY - Coredump will not include grid-scale shared memory for the warp that the dumped kernel belonged to.
- CU_COREDUMP_SKIP_LOCAL_MEMORY - Coredump will not include local memory from the kernel.
- CU_COREDUMP_LIGHTWEIGHT_FLAGS - Enables all of the above options. Equiavlent to setting the CU_COREDUMP_LIGHTWEIGHT attribute to ::true.
- CU_COREDUMP_SKIP_ABORT - If set, GPU exceptions will not raise an abort() in the host CPU process. Same functional goal as CU_COREDUMP_TRIGGER_HOST but better reflects the default behavior.
- CU_COREDUMP_SKIP_CONSTBANK_MEMORY - Coredump will not include constbank memory.
- CU_COREDUMP_SKIP_TENSOR_MEMORY - Coredump will not include tensor memory data. Tensor memory metadata may still be included.
- CU_COREDUMP_GZIP_COMPRESS - The generated coredump will be compressed with gzip, and .gz suffix will be appended to the filename, if it’s not a part of it already.
- CU_COREDUMP_FAULTED_CONTEXTS_ONLY - The coredump will only include contexts that have encountered an exception or a trap.
- CU_COREDUMP_NO_ERRBAR_AT_EXIT - By default, when coredumps are requested, the GPU will ensure memory faults and other errors prevent warps from exiting, if possible. This can potentially affect the performance of the application. Setting this flag will disable this functionality, making it possible for faulted warps to exit, but also avoiding the potential performance hit.
- CU_COREDUMP_LOG_ONLY - Setting this flag will disable actual generation of the coredump file, but exception details will still be logged.

Parameters

attrib – - The enum defining which value to fetch.
value – - void* containing the requested data.
size – - The size of the memory region value points to.

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_NOT_PERMITTED, CUDA_ERROR_DEINITIALIZED, CUDA_ERROR_NOT_INITIALIZED, CUDA_ERROR_INVALID_CONTEXT, CUDA_ERROR_CONTEXT_IS_DESTROYED

CUresult cuCoredumpGetAttributeGlobal(CUcoredumpSettings attrib, void *value, size_t *size)

Allows caller to fetch a coredump attribute value for the entire application.

Returns in *value the requested value specified by attrib. It is up to the caller to ensure that the data type and size of *value matches the request.

If the caller calls this function with *value equal to NULL, the size of the memory region (in bytes) expected for attrib will be placed in size.

The supported attributes are:

CU_COREDUMP_ENABLE_ON_EXCEPTION: Bool where ::true means that GPU exceptions from this context will create a coredump at the location specified by CU_COREDUMP_FILE. The default value is ::false.
CU_COREDUMP_TRIGGER_HOST: Bool where ::true means that the host CPU will also create a coredump. The default value is ::true unless set to ::false globally or or locally. This value is deprecated as of CUDA 12.5 - raise the CU_COREDUMP_SKIP_ABORT flag to disable host device abort() if needed.
CU_COREDUMP_LIGHTWEIGHT: Bool where ::true means that any resulting coredumps will not have a dump of GPU memory or non-reloc ELF images. The default value is ::false. This attribute is deprecated as of CUDA 12.5, please use CU_COREDUMP_GENERATION_FLAGS instead.
CU_COREDUMP_ENABLE_USER_TRIGGER: Bool where ::true means that a coredump can be created by writing to the system pipe specified by CU_COREDUMP_PIPE. The default value is ::false.
CU_COREDUMP_FILE: String of up to 1023 characters that defines the location where any coredumps generated by this context will be written. The default value is ::core.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA applications and ::PID is the process ID of the CUDA application.
CU_COREDUMP_PIPE: String of up to 1023 characters that defines the name of the pipe that will be monitored if user-triggered coredumps are enabled. The default value is ::corepipe.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA application and ::PID is the process ID of the CUDA application.
CU_COREDUMP_GENERATION_FLAGS: An integer with values to allow granular control the data contained in a coredump specified as a bitwise OR combination of the following values:
- CU_COREDUMP_DEFAULT_FLAGS - if set by itself, coredump generation returns to its default settings of including all memory regions that it is able to access
- CU_COREDUMP_SKIP_NONRELOCATED_ELF_IMAGES - Coredump will not include the data from CUDA source modules that are not relocated at runtime.
- CU_COREDUMP_SKIP_GLOBAL_MEMORY - Coredump will not include device-side global data that does not belong to any context.
- CU_COREDUMP_SKIP_SHARED_MEMORY - Coredump will not include grid-scale shared memory for the warp that the dumped kernel belonged to.
- CU_COREDUMP_SKIP_LOCAL_MEMORY - Coredump will not include local memory from the kernel.
- CU_COREDUMP_LIGHTWEIGHT_FLAGS - Enables all of the above options. Equiavlent to setting the CU_COREDUMP_LIGHTWEIGHT attribute to ::true.
- CU_COREDUMP_SKIP_ABORT - If set, GPU exceptions will not raise an abort() in the host CPU process. Same functional goal as CU_COREDUMP_TRIGGER_HOST but better reflects the default behavior.
- CU_COREDUMP_SKIP_CONSTBANK_MEMORY - Coredump will not include constbank memory.
- CU_COREDUMP_SKIP_TENSOR_MEMORY - Coredump will not include tensor memory data. Tensor memory metadata may still be included.
- CU_COREDUMP_GZIP_COMPRESS - The generated coredump will be compressed with gzip, and .gz suffix will be appended to the filename, if it’s not a part of it already.
- CU_COREDUMP_FAULTED_CONTEXTS_ONLY - The coredump will only include contexts that have encountered an exception or a trap.
- CU_COREDUMP_NO_ERRBAR_AT_EXIT - By default, when coredumps are requested, the GPU will ensure memory faults and other errors prevent warps from exiting, if possible. This can potentially affect the performance of the application. Setting this flag will disable this functionality, making it possible for faulted warps to exit, but also avoiding the potential performance hit.
- CU_COREDUMP_LOG_ONLY - Setting this flag will disable actual generation of the coredump file, but exception details will still be logged.

Parameters

attrib – - The enum defining which value to fetch.
value – - void* containing the requested data.
size – - The size of the memory region value points to.

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE

CUresult cuCoredumpRegisterCompleteCallback(CUcoredumpStatusCallback callback, void *userData, CUcoredumpCallbackHandle *callbackOut)

Register a callback to be invoked when a GPU coredump completes.

This function registers a callback that will be called when a GPU coredump has been fully collected and written to disk. Callbacks are executed in the order they were registered. The same callback function can be registered multiple times with different userData, and each registration will receive a unique handle.

Note

Callbacks execute synchronously during the coredump process and will block coredump progress while running.

Parameters

callback – - The callback function to register
userData – - User data pointer to pass to the callback
callbackOut – - Location to store the callback handle (optional, may be NULL)

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_OUT_OF_MEMORY

CUresult cuCoredumpRegisterStartCallback(CUcoredumpStatusCallback callback, void *userData, CUcoredumpCallbackHandle *callbackOut)

Register a callback to be invoked when a GPU coredump begins.

This function registers a callback that will be called when a GPU coredump is initiated, before any coredump data is collected. Callbacks are executed in the order they were registered. The same callback function can be registered multiple times with different userData, and each registration will receive a unique handle.

Note

Callbacks execute synchronously during the coredump process and will block coredump progress while running.

Parameters

callback – - The callback function to register
userData – - User data pointer to pass to the callback
callbackOut – - Location to store the callback handle (optional, may be NULL)

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_OUT_OF_MEMORY

CUresult cuCoredumpSetAttribute(CUcoredumpSettings attrib, void *value, size_t *size)

Allows caller to set a coredump attribute value for the current context.

This function should be considered an alternate interface to the CUDA-GDB environment variables defined in this document: https://docs.nvidia.com/cuda/cuda-gdb/index.html#gpu-coredump

An important design decision to note is that any coredump environment variable values set before CUDA initializes will take permanent precedence over any values set with this function. This decision was made to ensure no change in behavior for any users that may be currently using these variables to get coredumps.

*value shall contain the requested value specified by set. It is up to the caller to ensure that the data type and size of *value matches the request.

If the caller calls this function with *value equal to NULL, the size of the memory region (in bytes) expected for set will be placed in size.

/note This function will return CUDA_ERROR_NOT_SUPPORTED if the caller attempts to set CU_COREDUMP_ENABLE_ON_EXCEPTION on a GPU of with Compute Capability < 6.0. cuCoredumpSetAttributeGlobal works on those platforms as an alternative.

/note CU_COREDUMP_ENABLE_USER_TRIGGER and CU_COREDUMP_PIPE cannot be set on a per-context basis.

The supported attributes are:

CU_COREDUMP_ENABLE_ON_EXCEPTION: Bool where ::true means that GPU exceptions from this context will create a coredump at the location specified by CU_COREDUMP_FILE. The default value is ::false.
CU_COREDUMP_TRIGGER_HOST: Bool where ::true means that the host CPU will also create a coredump. The default value is ::true unless set to ::false globally or or locally. This value is deprecated as of CUDA 12.5 - raise the CU_COREDUMP_SKIP_ABORT flag to disable host device abort() if needed.
CU_COREDUMP_LIGHTWEIGHT: Bool where ::true means that any resulting coredumps will not have a dump of GPU memory or non-reloc ELF images. The default value is ::false. This attribute is deprecated as of CUDA 12.5, please use CU_COREDUMP_GENERATION_FLAGS instead.
CU_COREDUMP_FILE: String of up to 1023 characters that defines the location where any coredumps generated by this context will be written. The default value is ::core.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA applications and ::PID is the process ID of the CUDA application.
CU_COREDUMP_GENERATION_FLAGS: An integer with values to allow granular control the data contained in a coredump specified as a bitwise OR combination of the following values:
- CU_COREDUMP_DEFAULT_FLAGS - if set by itself, coredump generation returns to its default settings of including all memory regions that it is able to access
- CU_COREDUMP_SKIP_NONRELOCATED_ELF_IMAGES - Coredump will not include the data from CUDA source modules that are not relocated at runtime.
- CU_COREDUMP_SKIP_GLOBAL_MEMORY - Coredump will not include device-side global data that does not belong to any context.
- CU_COREDUMP_SKIP_SHARED_MEMORY - Coredump will not include grid-scale shared memory for the warp that the dumped kernel belonged to.
- CU_COREDUMP_SKIP_LOCAL_MEMORY - Coredump will not include local memory from the kernel.
- CU_COREDUMP_LIGHTWEIGHT_FLAGS - Enables all of the above options. Equiavlent to setting the CU_COREDUMP_LIGHTWEIGHT attribute to ::true.
- CU_COREDUMP_SKIP_ABORT - If set, GPU exceptions will not raise an abort() in the host CPU process. Same functional goal as CU_COREDUMP_TRIGGER_HOST but better reflects the default behavior.
- CU_COREDUMP_SKIP_CONSTBANK_MEMORY - Coredump will not include constbank memory.
- CU_COREDUMP_SKIP_TENSOR_MEMORY - Coredump will not include tensor memory data. Tensor memory metadata may still be included.
- CU_COREDUMP_GZIP_COMPRESS - The generated coredump will be compressed with gzip, and .gz suffix will be appended to the filename, if it’s not a part of it already.
- CU_COREDUMP_FAULTED_CONTEXTS_ONLY - The coredump will only include contexts that have encountered an exception or a trap.
- CU_COREDUMP_NO_ERRBAR_AT_EXIT - By default, when coredumps are requested, the GPU will ensure memory faults and other errors prevent warps from exiting, if possible. This can potentially affect the performance of the application. Setting this flag will disable this functionality, making it possible for faulted warps to exit, but also avoiding the potential performance hit.
- CU_COREDUMP_LOG_ONLY - Setting this flag will disable actual generation of the coredump file, but exception details will still be logged.

Note

CU_COREDUMP_GENERATION_FLAGS replaces all previously set coredump flags. Mixing CU_COREDUMP_GENERATION_FLAGS with the deprecated boolean attributes (CU_COREDUMP_TRIGGER_HOST, CU_COREDUMP_LIGHTWEIGHT) can result in undefined behavior. To avoid issues, either use only CU_COREDUMP_GENERATION_FLAGS or combine all desired flag bits (including CU_COREDUMP_SKIP_ABORT) in a single call.

Parameters

attrib – - The enum defining which value to set.
value – - void* containing the requested data.
size – - The size of the memory region value points to.

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_NOT_PERMITTED, CUDA_ERROR_DEINITIALIZED, CUDA_ERROR_NOT_INITIALIZED, CUDA_ERROR_INVALID_CONTEXT, CUDA_ERROR_CONTEXT_IS_DESTROYED, CUDA_ERROR_NOT_SUPPORTED

CUresult cuCoredumpSetAttributeGlobal(CUcoredumpSettings attrib, void *value, size_t *size)

Allows caller to set a coredump attribute value globally.

This function should be considered an alternate interface to the CUDA-GDB environment variables defined in this document: https://docs.nvidia.com/cuda/cuda-gdb/index.html#gpu-coredump

An important design decision to note is that any coredump environment variable values set before CUDA initializes will take permanent precedence over any values set with this function. This decision was made to ensure no change in behavior for any users that may be currently using these variables to get coredumps.

*value shall contain the requested value specified by set. It is up to the caller to ensure that the data type and size of *value matches the request.

If the caller calls this function with *value equal to NULL, the size of the memory region (in bytes) expected for set will be placed in size.

The supported attributes are:

CU_COREDUMP_ENABLE_ON_EXCEPTION: Bool where ::true means that GPU exceptions from this context will create a coredump at the location specified by CU_COREDUMP_FILE. The default value is ::false.
CU_COREDUMP_TRIGGER_HOST: Bool where ::true means that the host CPU will also create a coredump. The default value is ::true unless set to ::false globally or or locally. This value is deprecated as of CUDA 12.5 - raise the CU_COREDUMP_SKIP_ABORT flag to disable host device abort() if needed.
CU_COREDUMP_LIGHTWEIGHT: Bool where ::true means that any resulting coredumps will not have a dump of GPU memory or non-reloc ELF images. The default value is ::false. This attribute is deprecated as of CUDA 12.5, please use CU_COREDUMP_GENERATION_FLAGS instead.
CU_COREDUMP_ENABLE_USER_TRIGGER: Bool where ::true means that a coredump can be created by writing to the system pipe specified by CU_COREDUMP_PIPE. The default value is ::false.
CU_COREDUMP_FILE: String of up to 1023 characters that defines the location where any coredumps generated by this context will be written. The default value is ::core.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA applications and ::PID is the process ID of the CUDA application.
CU_COREDUMP_PIPE: String of up to 1023 characters that defines the name of the pipe that will be monitored if user-triggered coredumps are enabled. This value may not be changed after CU_COREDUMP_ENABLE_USER_TRIGGER is set to ::true. The default value is ::corepipe.cuda.HOSTNAME.PID where ::HOSTNAME is the host name of the machine running the CUDA application and ::PID is the process ID of the CUDA application.
CU_COREDUMP_GENERATION_FLAGS: An integer with values to allow granular control the data contained in a coredump specified as a bitwise OR combination of the following values:
- CU_COREDUMP_DEFAULT_FLAGS - if set by itself, coredump generation returns to its default settings of including all memory regions that it is able to access
- CU_COREDUMP_SKIP_NONRELOCATED_ELF_IMAGES - Coredump will not include the data from CUDA source modules that are not relocated at runtime.
- CU_COREDUMP_SKIP_GLOBAL_MEMORY - Coredump will not include device-side global data that does not belong to any context.
- CU_COREDUMP_SKIP_SHARED_MEMORY - Coredump will not include grid-scale shared memory for the warp that the dumped kernel belonged to.
- CU_COREDUMP_SKIP_LOCAL_MEMORY - Coredump will not include local memory from the kernel.
- CU_COREDUMP_LIGHTWEIGHT_FLAGS - Enables all of the above options. Equiavlent to setting the CU_COREDUMP_LIGHTWEIGHT attribute to ::true.
- CU_COREDUMP_SKIP_ABORT - If set, GPU exceptions will not raise an abort() in the host CPU process. Same functional goal as CU_COREDUMP_TRIGGER_HOST but better reflects the default behavior.
- CU_COREDUMP_SKIP_CONSTBANK_MEMORY - Coredump will not include constbank memory.
- CU_COREDUMP_SKIP_TENSOR_MEMORY - Coredump will not include tensor memory data. Tensor memory metadata may still be included.
- CU_COREDUMP_GZIP_COMPRESS - The generated coredump will be compressed with gzip, and .gz suffix will be appended to the filename, if it’s not a part of it already.
- CU_COREDUMP_FAULTED_CONTEXTS_ONLY - The coredump will only include contexts that have encountered an exception or a trap.
- CU_COREDUMP_NO_ERRBAR_AT_EXIT - By default, when coredumps are requested, the GPU will ensure memory faults and other errors prevent warps from exiting, if possible. This can potentially affect the performance of the application. Setting this flag will disable this functionality, making it possible for faulted warps to exit, but also avoiding the potential performance hit.
- CU_COREDUMP_LOG_ONLY - Setting this flag will disable actual generation of the coredump file, but exception details will still be logged.

Note

CU_COREDUMP_GENERATION_FLAGS replaces all previously set coredump flags. Mixing CU_COREDUMP_GENERATION_FLAGS with the deprecated boolean attributes (CU_COREDUMP_TRIGGER_HOST, CU_COREDUMP_LIGHTWEIGHT) can result in undefined behavior. To avoid issues, either use only CU_COREDUMP_GENERATION_FLAGS or combine all desired flag bits (including CU_COREDUMP_SKIP_ABORT) in a single call.

Parameters

attrib – - The enum defining which value to set.
value – - void* containing the requested data.
size – - The size of the memory region value points to.

Returns

CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_NOT_PERMITTED

6.4.3. Typedefs

typedef struct CUcoredumpCallbackEntry_st *CUcoredumpCallbackHandle

Opaque handle representing a registered coredump status callback.

This handle is returned when registering a callback and must be provided when deregistering the callback.

typedef void (*CUcoredumpStatusCallback)(void *userData, int pid, CUdevice dev)

Callback function prototype for GPU coredump status notifications.

This callback will be invoked when a GPU coredump begins or completes, depending on which registration function was used. The callback executes synchronously during the coredump process.

Param userData: - User-provided data pointer that was passed during registration
Param pid: - Process ID of the faulted process
Param dev: - Device ordinal where the coredump occurred