| #
cda3d176 |
| 03-Aug-2023 |
Levend Sayar <levendsayar@gmail.com> |
gpu/cuda: fix build with external GDRCopy
If gdrapi.h is available, meson sets DRIVERS_GPU_CUDA_GDRCOPY_H as 1.
This causes gdrcopy.c build to give an error;
because compiler can not find signature
gpu/cuda: fix build with external GDRCopy
If gdrapi.h is available, meson sets DRIVERS_GPU_CUDA_GDRCOPY_H as 1.
This causes gdrcopy.c build to give an error;
because compiler can not find signature of getenv.
stdlib.h is included for the definition of getenv function.
Bugzilla ID: 1133
Fixes: ca12f5e8a7db ("gpu/cuda: mark unused GDRCopy functions parameters")
Cc: stable@dpdk.org
Reported-by: John Romein <romein@astron.nl>
Signed-off-by: Levend Sayar <levendsayar@gmail.com>
Acked-by: Elena Agostini <eagostini@nvidia.com>
show more ...
|
| #
ca12f5e8 |
| 10-Mar-2022 |
Elena Agostini <eagostini@nvidia.com> |
gpu/cuda: mark unused GDRCopy functions parameters
Without enabling gdrcopy, DPDK build has warnings about
unused gdrcopy functions input parameters.
This patch set them to __rte_unused .
Fixes: 2
gpu/cuda: mark unused GDRCopy functions parameters
Without enabling gdrcopy, DPDK build has warnings about
unused gdrcopy functions input parameters.
This patch set them to __rte_unused .
Fixes: 24c7759 ("gpu/cuda: map GPU memory with GDRCopy")
Signed-off-by: Elena Agostini <eagostini@nvidia.com>
show more ...
|
| #
24c77594 |
| 25-Feb-2022 |
Elena Agostini <eagostini@nvidia.com> |
gpu/cuda: map GPU memory with GDRCopy
To enable the gpudev rte_gpu_mem_cpu_map feature to expose
GPU memory to the CPU, the GPU CUDA driver library needs
the GDRCopy library and driver.
If DPDK is
gpu/cuda: map GPU memory with GDRCopy
To enable the gpudev rte_gpu_mem_cpu_map feature to expose
GPU memory to the CPU, the GPU CUDA driver library needs
the GDRCopy library and driver.
If DPDK is built without GDRCopy, the GPU CUDA driver returns
error if the is invoked rte_gpu_mem_cpu_map.
All the others GPU CUDA driver functionalities are not affected by
the absence of GDRCopy, thus this is an optional functionality
that can be enabled in the GPU CUDA driver.
CUDA driver documentation has been updated accordingly.
Signed-off-by: Elena Agostini <eagostini@nvidia.com>
show more ...
|