Skip to content

Fix OOM state-desync in generated array-member setters #2414

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
rparolin wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix OOM state-desync in generated array-member setters #2414

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 15 additions & 15 deletions cuda_bindings/cuda/bindings/cyruntime.pxd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
#
# This code was automatically generated across versions from 12.9.0 to 13.3.0. Do not modify it directly.

# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=2b62a3b56226f5cf6953f578057af792369a100b7775328c5aa66d81d780d2ac
# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=a7205ec1f1749acd8f9e32e85660d77d8ccb60a0d038bb63fb4b015496f23d07
from libc.stdint cimport uint32_t, uint64_t


Expand Down Expand Up @@ -1254,6 +1254,20 @@ cdef extern from 'library_types.h':
CUDA_EMULATION_STRATEGY_EAGER
ctypedef cudaEmulationStrategy_t cudaEmulationStrategy

cdef extern from 'library_types.h':

@mdboom mdboom Jul 23, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks like maybe there is a non-deterministic ordering bug in the generator here. Not to fix now, but I will file a bug over there.

cdef enum cudaEmulationMantissaControl_t:
CUDA_EMULATION_MANTISSA_CONTROL_DYNAMIC
CUDA_EMULATION_MANTISSA_CONTROL_FIXED
ctypedef cudaEmulationMantissaControl_t cudaEmulationMantissaControl

cdef extern from 'library_types.h':
cdef enum cudaEmulationSpecialValuesSupport_t:
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_DEFAULT
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_NONE
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_INFINITY
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_NAN
ctypedef cudaEmulationSpecialValuesSupport_t cudaEmulationSpecialValuesSupport

cdef extern from 'driver_types.h':
cdef enum cudaDevSmResourceGroup_flags:
cudaDevSmResourceGroupDefault
Expand All @@ -1276,20 +1290,6 @@ cdef extern from 'driver_types.h':
cudaDevWorkqueueConfigScopeDeviceCtx
cudaDevWorkqueueConfigScopeGreenCtxBalanced

cdef extern from 'library_types.h':
cdef enum cudaEmulationMantissaControl_t:
CUDA_EMULATION_MANTISSA_CONTROL_DYNAMIC
CUDA_EMULATION_MANTISSA_CONTROL_FIXED
ctypedef cudaEmulationMantissaControl_t cudaEmulationMantissaControl

cdef extern from 'library_types.h':
cdef enum cudaEmulationSpecialValuesSupport_t:
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_DEFAULT
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_NONE
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_INFINITY
CUDA_EMULATION_SPECIAL_VALUES_SUPPORT_NAN
ctypedef cudaEmulationSpecialValuesSupport_t cudaEmulationSpecialValuesSupport

cdef extern from 'driver_types.h':
cdef enum cudaHostTaskSyncMode:
cudaHostTaskBlocking
Expand Down
321 changes: 223 additions & 98 deletions cuda_bindings/cuda/bindings/driver.pyx
View file
Open in desktop

Large diffs are not rendered by default.

14 changes: 4 additions & 10 deletions cuda_bindings/cuda/bindings/nvml.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
# SPDX-License-Identifier: Apache-2.0
#
# This code was automatically generated across versions from 12.9.1 to 13.3.0. Do not modify it directly.
# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=e47a16bd9956de14a991ded5d1aef667cdd26a141e27db5b2015b91be6918d3c
# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=6dc4cfe2b034290ebc642bf79eea9ec031190b3a02615d5a2c8c0e6c26100bbb


# <<<< PREAMBLE CONTENT >>>>
Expand Down Expand Up @@ -840,7 +840,7 @@ class GpmMetricId(_cyb_FastEnum):
GPM_METRIC_HMMA_TENSOR_UTIL = (NVML_GPM_METRIC_HMMA_TENSOR_UTIL, "Percentage of time the GPU's SMs were doing HMMA tensor operations. 0.0 - 100.0.")
GPM_METRIC_DMMA_TENSOR_UTIL = (NVML_GPM_METRIC_DMMA_TENSOR_UTIL, "Percentage of time the GPU's SMs were doing DMMA tensor operations. 0.0 - 100.0.")
GPM_METRIC_IMMA_TENSOR_UTIL = (NVML_GPM_METRIC_IMMA_TENSOR_UTIL, "Percentage of time the GPU's SMs were doing IMMA tensor operations. 0.0 - 100.0.")
GPM_METRIC_DRAM_BW_UTIL = (NVML_GPM_METRIC_DRAM_BW_UTIL, 'Percentage of DRAM bw used vs theoretical maximum. 0.0 - 100.0 *\u200d/.')
GPM_METRIC_DRAM_BW_UTIL = (NVML_GPM_METRIC_DRAM_BW_UTIL, 'Percentage of DRAM bw used vs theoretical maximum. `0.0 - 100.0 */`.')
GPM_METRIC_FP64_UTIL = (NVML_GPM_METRIC_FP64_UTIL, "Percentage of time the GPU's SMs were doing non-tensor FP64 math. 0.0 - 100.0.")
GPM_METRIC_FP32_UTIL = (NVML_GPM_METRIC_FP32_UTIL, "Percentage of time the GPU's SMs were doing non-tensor FP32 math. 0.0 - 100.0.")
GPM_METRIC_FP16_UTIL = (NVML_GPM_METRIC_FP16_UTIL, "Percentage of time the GPU's SMs were doing non-tensor FP16 math. 0.0 - 100.0.")
Expand Down Expand Up @@ -25091,7 +25091,7 @@ cpdef int device_get_virtualization_mode(intptr_t device) except? -1:
device (intptr_t): Identifier of the target device.

Returns:
int: Reference to virtualization mode. One of NVML_GPU_VIRTUALIZATION_?.
int: Reference to virtualization mode. One of ``NVML_GPU_VIRTUALIZATION_?``.

.. seealso:: `nvmlDeviceGetVirtualizationMode`
"""
Expand Down Expand Up @@ -25125,7 +25125,7 @@ cpdef device_set_virtualization_mode(intptr_t device, int virtual_mode):

Args:
device (intptr_t): Identifier of the target device.
virtual_mode (GpuVirtualizationMode): virtualization mode. One of NVML_GPU_VIRTUALIZATION_?.
virtual_mode (GpuVirtualizationMode): virtualization mode. One of ``NVML_GPU_VIRTUALIZATION_?``.

.. seealso:: `nvmlDeviceSetVirtualizationMode`
"""
Expand Down Expand Up @@ -28464,10 +28464,4 @@ cpdef str vgpu_type_get_name(unsigned int vgpu_type_id):
__status__ = nvmlVgpuTypeGetName(<nvmlVgpuTypeId_t>vgpu_type_id, vgpu_type_name, <unsigned int*>size)
check_status(__status__)
return cpython.PyUnicode_FromStringAndSize(vgpu_type_name, size[0])


# Cleanup some docstrings that don't parse as rst.
device_get_virtualization_mode.__doc__ = device_get_virtualization_mode.__doc__.replace("NVML_GPU_VIRTUALIZATION_?", "``NVML_GPU_VIRTUALIZATION_?``")
device_set_virtualization_mode.__doc__ = device_set_virtualization_mode.__doc__.replace("NVML_GPU_VIRTUALIZATION_?", "``NVML_GPU_VIRTUALIZATION_?``")
GpmMetricId.GPM_METRIC_DRAM_BW_UTIL.__doc__ = "Percentage of DRAM bw used vs theoretical maximum. ``0.0 - 100.0 *\u200d/``."
del _cyb_FastEnum
38 changes: 19 additions & 19 deletions cuda_bindings/cuda/bindings/runtime.pxd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
# SPDX-License-Identifier: Apache-2.0

# This code was automatically generated with version 13.3.0. Do not modify it directly.
# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=0f2431380680008795336b7acb5ccd83dba7a6e05d0c81c2b5f825bc95576ccd
# CYTHON-BINDINGS-GENERATED-DO-NOT-MODIFY-THIS-FILE: format=1; content-sha256=ea44514ba11796759754ddfb1ae93c9b7e1d510a09f7349a82a2201dd0c0c159
cimport cuda.bindings.cyruntime as cyruntime

include "_lib/utils.pxd"
Expand Down Expand Up @@ -4023,12 +4023,12 @@ cdef class cudaLaunchAttributeValue:
with the following fields: - `x` - The X dimension of the preferred
cluster, in blocks. Must be a divisor of the grid X dimension, and
must be a multiple of the `x` field of
::cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension
of the preferred cluster, in blocks. Must be a divisor of the grid
Y dimension, and must be a multiple of the `y` field of
::cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension
of the preferred cluster, in blocks. Must be equal to the `z` field
of ::cudaLaunchAttributeValue::clusterDim.
cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension of
the preferred cluster, in blocks. Must be a divisor of the grid Y
dimension, and must be a multiple of the `y` field of
cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension of
the preferred cluster, in blocks. Must be equal to the `z` field of
cudaLaunchAttributeValue::clusterDim.


launchCompletionEvent : anon_struct20
Expand Down Expand Up @@ -4830,12 +4830,12 @@ cdef class cudaStreamAttrValue(cudaLaunchAttributeValue):
with the following fields: - `x` - The X dimension of the preferred
cluster, in blocks. Must be a divisor of the grid X dimension, and
must be a multiple of the `x` field of
::cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension
of the preferred cluster, in blocks. Must be a divisor of the grid
Y dimension, and must be a multiple of the `y` field of
::cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension
of the preferred cluster, in blocks. Must be equal to the `z` field
of ::cudaLaunchAttributeValue::clusterDim.
cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension of
the preferred cluster, in blocks. Must be a divisor of the grid Y
dimension, and must be a multiple of the `y` field of
cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension of
the preferred cluster, in blocks. Must be equal to the `z` field of
cudaLaunchAttributeValue::clusterDim.


launchCompletionEvent : anon_struct20
Expand Down Expand Up @@ -4959,12 +4959,12 @@ cdef class cudaKernelNodeAttrValue(cudaLaunchAttributeValue):
with the following fields: - `x` - The X dimension of the preferred
cluster, in blocks. Must be a divisor of the grid X dimension, and
must be a multiple of the `x` field of
::cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension
of the preferred cluster, in blocks. Must be a divisor of the grid
Y dimension, and must be a multiple of the `y` field of
::cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension
of the preferred cluster, in blocks. Must be equal to the `z` field
of ::cudaLaunchAttributeValue::clusterDim.
cudaLaunchAttributeValue::clusterDim. - `y` - The Y dimension of
the preferred cluster, in blocks. Must be a divisor of the grid Y
dimension, and must be a multiple of the `y` field of
cudaLaunchAttributeValue::clusterDim. - `z` - The Z dimension of
the preferred cluster, in blocks. Must be equal to the `z` field of
cudaLaunchAttributeValue::clusterDim.


launchCompletionEvent : anon_struct20
Expand Down
Loading
Loading