limingqi107
|
44959f8874
|
gpu kernel runtime code review
|
5 years ago |
lizhenyu
|
1b4a7cdeb7
|
fix mem swap bug
|
5 years ago |
limingqi107
|
0f4397cece
|
fix all nop node graph execute
|
5 years ago |
mindspore-ci-bot
|
4642df207a
|
!2210 gpu optimize the max device memory config
Merge pull request !2210 from limingqi107/max_device_memory_optimize
|
5 years ago |
limingqi107
|
55b3557c0d
|
gpu optimize the max device memory config
|
5 years ago |
limingqi107
|
20083679a0
|
gpu memreuse supports summary node
|
5 years ago |
limingqi107
|
da2b93916e
|
gpu device code review
|
5 years ago |
lizhenyu
|
0a55ebf6e9
|
fix code review defects
|
5 years ago |
lizhenyu
|
fa4d9b2846
|
fix resource release bug of memory swap
|
5 years ago |
lizhenyu
|
23a57476da
|
change gpu kernel runtime to support memory swap
|
5 years ago |
caifubi
|
5b963aef2b
|
Change uintptr_t to void ptr
|
5 years ago |
limingqi107
|
d9197b591a
|
gpu optimize the use of reference count
|
5 years ago |
limingqi107
|
63f3a2caac
|
gpu optimize some return values of dynamic memory pool
|
5 years ago |
lizhenyu
|
3a4c28fa33
|
change directory of akg cuda kernel
|
5 years ago |
limingqi107
|
664f2628e5
|
optimize gpu allReduce alloc memory performance
|
5 years ago |
limingqi107
|
0f0e8fe874
|
gpu dynamic memory pool can not reuse allReduce in multi-stream
|
5 years ago |
limingqi107
|
2891f0d20d
|
gpu dynamic memory pool supports multi-allReduce
|
5 years ago |
limingqi107
|
5e01b94ccd
|
gpu dynamic memory pool suppoerts multi-graph
|
5 years ago |
limingqi107
|
99f12f9105
|
gpu uses dynamic memory pool by default
|
5 years ago |
kswang
|
b8a7e73f7d
|
fix reshape output and clearres error
|
6 years ago |
kswang
|
bef62db128
|
add ascend mem pool
|
6 years ago |
kswang
|
fb343bd607
|
add mem manager
|
6 years ago |
zhunaipan
|
930a1fb0a8
|
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
|
6 years ago |