Skip to content

P
Paddle

PaddlePaddle / Paddle
接近 2 年 前同步成功

通知 2323
Star 20933
Fork 5424
P
Paddle
“ecfe1c788a04fa7f24c754ed77ddf50db7049f07”上不存在“components/net/lwip-2.0.2/src/lwipopts.h”
提交 bcf6f828 编写于 作者: Z zhouyingfeng
浏览文件
操作

add chinese doc for gpu-profiling, and fix the out-dated lineNos in en doc... 

add chinese doc for gpu-profiling, and fix the out-dated lineNos in en doc referenced to a code file.

resolve #834
上级 71b42039
隐藏空白更改
内联 并排
Showing with 299 addition and 60 deletion
doc/howto/optimization/gpu_profiling_cn.rst 0 → 100644
浏览文件 @ bcf6f828
PaddlePaddle 中的性能分析
=====================================
此教程将向您分步介绍如何使用内置的定时工具、 **nvprof** 或 **nvvp** 来运行性能分析和调优。
- 什么是性能分析？
- 为什么需要性能分析？
- 如何进行性能分析？
- 性能分析工具介绍
- 详细教程
- 性能分析小技巧
什么是性能分析？
================
在软件工程的范畴里，性能分析（Profiling）是一个动态程序分析的术语，它可以指测量一个程序的空间（内存）复杂度或时间复杂度，
也可以说是某些特定指令的使用情况，或者是函数调用的频率和耗时等。通常情况下，分析得到的信息用于协助进行程序的优化。
简单来说，性能分析工具是用于给应用程序的性能做定量分析的。如果想很好的理解程序的行为，那程序分析工具是必不可少的利器。简单的性能分析，可以告诉您某个操作到底花了多长时间？而更深入的分析，甚至能解释为什么某个操作花了很长时间？
为什么需要性能分析？
============================
训练好一个深层神经网络通常要耗费非常长的时间，所以性能也就逐步变成了深度学习领域最重要的指标。
而优化性能的首要任务，是需要了解哪些步骤拖慢了整体。
如果某一块根本就不怎么耗时，那也就不需要急着优化性能啦！
如何进行性能分析？
========================
为了达到性能最优，您可以采用下面五个步骤：
- 对代码进行性能分析
- 找到运行慢的部分
- 找到运行慢的原因
- 修改成更快的版本
- 再次对代码进行性能分析
Usually, processor has two key performance limits include float point throughput and
memory throughput. For GPU, it also need more parallelism to fulfill its potential.
This is why they can be so fast.
通常情况下，处理器有两个关键性能限制：一个是浮点计算量，另一个是内存操作量。
GPU则还需要高并行性，才能发挥其全部能力。这正是它们速度快的原因。
性能分析工具介绍
======================
就通常的GPU性能分析来说，市面上已经有NVIDIA或第三方提供的众多工具。
**nvprof** 是Nvidia性能分析工具， **nvvp** 则是带GUI的Nvidia可视化性能分析工具。
在这个教程中，我们主要会介绍nvprof和nvvp。
:code:`test_GpuProfiler` from :code:`paddle/math/tests` directory will be used to evaluate
above profilers.
:code:`paddle/math/test` 目录中的 :code:`test_GpuProfiler` 就是用于展示上述分析工具的用法。
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++
:lines: 137-151
:linenos:
上述的代码片段包含了两种方法，您可以任意使用一个或两个来对感兴趣的代码段做性能分析。
1. :code:`REGISTER_TIMER_INFO` 是一个内置的定时器封装，可以用来计算CPU函数或cuda内核的时间消耗。
2. :code:`REGISTER_GPU_PROFILER` is a general purpose wrapper object of :code:`cudaProfilerStart` and :code:`cudaProfilerStop` to avoid
program crashes when CPU version of PaddlePaddle invokes them.
3. :code:`REGISTER_GPU_PROFILER` 是一个封装对象，封装了 :code:`cudaProfilerStart` 和 :code:`cudaProfileStop` 两个操作；同时其内部实现可以避免纯CPU版本PaddlePaddle在执行本语句时发生崩溃。
您会在接下来的部分中获得更多的细节介绍。
详细教程
============
内置定时器
------------
如果想要启用PaddlePaddle的内置定时器，您首先需要在相关代码段中加入 :code:`REGISTER_TIMER_INFO`。
接下来就可以使用 :code:`printStatus` 或者 :code:`printAllStatus` 函数来将信息输出到界面中。
下面举个简单的例子：
1. 加入 :code:`REGISTER_TIMER_INFO` 和 :code:`printAllStatus` 函数（如高亮部分）。
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++
:lines: 137-151
:emphasize-lines: 8-12,14
:linenos:
2. cmake配置中将 **WITH_TIMER** 打开，重新编译PaddlePaddle。
.. code-block:: bash
cmake .. -DWITH_TIMER=ON
make
3. 执行您的代码，并观察结果(如高亮部分）。
.. code-block:: bash
:emphasize-lines: 1,12-15
> ./paddle/math/tests/test_GpuProfiler
I1117 11:13:42.313065 2522362816 Util.cpp:155] commandline: ./paddle/math/tests/test_GpuProfiler
I1117 11:13:42.845065 2522362816 Util.cpp:130] Calling runInitFunctions
I1117 11:13:42.845208 2522362816 Util.cpp:143] Call runInitFunctions done.
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from Profiler
[ RUN ] Profiler.BilinearFwdBwd
I1117 11:13:42.845310 2522362816 test_GpuProfiler.cpp:114] Enable GPU Profiler Stat: [testBilinearFwdBwd] "numSamples = 10, channels = 16, im
gSizeX = 64, imgSizeY = 64"
I1117 11:13:42.850154 2522362816 ThreadLocal.cpp:37] thread use undeterministic rand seed:20659751
I1117 11:13:42.981501 2522362816 Stat.cpp:130] ======= StatSet: [GlobalStatInfo] status ======
I1117 11:13:42.981539 2522362816 Stat.cpp:133] Stat=testBilinearFwdBwd total=136.141 avg=136.141 max=136.141 min=136.141 count=1
I1117 11:13:42.981572 2522362816 Stat.cpp:141] ======= BarrierStatSet status ======
I1117 11:13:42.981575 2522362816 Stat.cpp:154] --------------------------------------------------
[ OK ] Profiler.BilinearFwdBwd (136 ms)
[----------] 1 test from Profiler (136 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (136 ms total)
[ PASSED ] 1 test.
nvprof 工具
----------------
要使用命令行分析工具 **nvprof**，您按如下步骤操作即可：
1. 将 :code:`REGISTER_GPU_PROFILER` 函数加到代码中（参考强调部分）。
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++
:lines: 137-151
:emphasize-lines: 6-7
:linenos:
2. cmake中将 **WITH_PROFILER** 配置打开，重新编译PaddlePaddle。
.. code-block:: bash
cmake .. -DWITH_PROFILER=ON
make
3. 使用 **nvprof** 来分析执行文件。
.. code-block:: bash
nvprof ./paddle/math/tests/test_GpuProfiler
然后，您就能获得如下的分析结果：
.. code-block:: bash
==78544== Profiling application: ./paddle/math/tests/test_GpuProfiler
==78544== Profiling result:
Time(%) Time Calls Avg Min Max Name
27.60% 9.6305ms 5 1.9261ms 3.4560us 6.4035ms [CUDA memcpy HtoD]
26.07% 9.0957ms 1 9.0957ms 9.0957ms 9.0957ms KeBilinearInterpBw
23.78% 8.2977ms 1 8.2977ms 8.2977ms 8.2977ms KeBilinearInterpFw
22.55% 7.8661ms 2 3.9330ms 1.5798ms 6.2863ms [CUDA memcpy DtoH]
==78544== API calls:
Time(%) Time Calls Avg Min Max Name
46.85% 682.28ms 8 85.285ms 12.639us 682.03ms cudaStreamCreateWithFlags
39.83% 580.00ms 4 145.00ms 302ns 550.27ms cudaFree
9.82% 143.03ms 9 15.892ms 8.7090us 142.78ms cudaStreamCreate
1.23% 17.983ms 7 2.5690ms 23.210us 6.4563ms cudaMemcpy
1.23% 17.849ms 2 8.9247ms 8.4726ms 9.3768ms cudaStreamSynchronize
0.66% 9.5969ms 7 1.3710ms 288.43us 2.4279ms cudaHostAlloc
0.13% 1.9530ms 11 177.54us 7.6810us 591.06us cudaMalloc
0.07% 1.0424ms 8 130.30us 1.6970us 453.72us cudaGetDevice
0.04% 527.90us 40 13.197us 525ns 253.99us cudaEventCreateWithFlags
0.03% 435.73us 348 1.2520us 124ns 42.704us cuDeviceGetAttribute
0.03% 419.36us 1 419.36us 419.36us 419.36us cudaGetDeviceCount
0.02% 260.75us 2 130.38us 129.32us 131.43us cudaGetDeviceProperties
0.02% 222.32us 2 111.16us 106.94us 115.39us cudaLaunch
0.01% 214.06us 4 53.514us 28.586us 77.655us cuDeviceGetName
0.01% 115.45us 4 28.861us 9.8250us 44.526us cuDeviceTotalMem
0.01% 83.988us 4 20.997us 578ns 77.760us cudaSetDevice
0.00% 38.918us 1 38.918us 38.918us 38.918us cudaEventCreate
0.00% 34.573us 31 1.1150us 279ns 12.784us cudaDeviceGetAttribute
0.00% 17.767us 1 17.767us 17.767us 17.767us cudaProfilerStart
0.00% 15.228us 2 7.6140us 3.5460us 11.682us cudaConfigureCall
0.00% 14.536us 2 7.2680us 1.1490us 13.387us cudaGetLastError
0.00% 8.6080us 26 331ns 173ns 783ns cudaSetupArgument
0.00% 5.5470us 6 924ns 215ns 2.6780us cuDeviceGet
0.00% 5.4090us 6 901ns 328ns 3.3320us cuDeviceGetCount
0.00% 4.1770us 3 1.3920us 1.0630us 1.8300us cuDriverGetVersion
0.00% 3.4650us 3 1.1550us 1.0810us 1.2680us cuInit
0.00% 830ns 1 830ns 830ns 830ns cudaRuntimeGetVersion
nvvp 工具
--------------
如果想使用可视化的分析器 **nvvp**，您可以导入 :code:`nvprof -o ...` 的输出，或者从工具的界面里运行您的应用。
**备注: nvvp 也支持CPU的性能分析** (需在nvvp界面中选上才能开启）
.. image:: nvvp1.png
:align: center
:scale: 33%
从内核函数的角度， **nvvp** 可以精确说明一个长耗时操作的具体原因。
同时，如下图所示， **nvvp** 的内核block使用情况、register使用情况和共享内存使用情况能让我们对GPU的整体使用有更好的理解。
.. image:: nvvp2.png
:align: center
:scale: 33%
而从应用的角度， **nvvp** 可以帮您提供一些定位性能瓶颈的建议。
例如，下图中就展示了一些关于data movement和compute utilization的建议，为您做性能调优提供了方向。
.. image:: nvvp3.png
:align: center
:scale: 33%
.. image:: nvvp4.png
:align: center
:scale: 33%
性能分析小技巧
==================
- 开始阶段，从 **nvprof** 和 **nvvp** 的输出信息入手是个不错的选择。
- 接下来可以考虑下时间线的分析。
- 如果真想挖掘内核深处的某个秘密，您最好先确认：这一块的耗时比例真的太高，值得深入分析。
- 可能的情况下，试着让输出的分析数据和理论值对应。
1) 例如，如果我知道内核花了10ms来移动1GB数据，那我会期望分析工具统计到速度是100GB/s。
2) 若有不一致之处，很有可能实际应用就是没有按照您的预期情况运行。
- 了解您的硬件：如果您的GPU理论可以达到6 TFLOPs（6万亿次浮点运算每秒），而当前已经有5.5 TFLOPs了，那估计这里的潜力就没啥好挖的了……
性能分析是性能优化的关键一步。有的时候简简单单的改变就能在性能上产生明显的优化效果！
当然，具体情况因人而异。
参考资料
===========
Jeremy Appleyard, `GPU Profiling for Deep Learning <http://www.robots.ox.ac.uk/~seminars/seminars/Extra/2015_10_08_JeremyAppleyard.pdf>`_, 2015
doc/howto/optimization/gpu_profiling_en.rst
浏览文件 @ bcf6f828
...@@ -49,7 +49,7 @@ For general GPU profiling, a bunch of tools are provided from both NVIDIA and th ...@@ -49,7 +49,7 @@ For general GPU profiling, a bunch of tools are provided from both NVIDIA and th
In this tutorial, we will focus on nvprof and nvvp. In this tutorial, we will focus on nvprof and nvvp.
:code:`test_GpuProfiler` from :code:`paddle/math/tests` directory will be used to evaluate :code:`test_GpuProfiler` from :code:`paddle/math/tests` directory will be used to evaluate
above profilers. above profilers.
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp .. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++ :language: c++
...@@ -79,8 +79,8 @@ As a simple example, consider the following: ...@@ -79,8 +79,8 @@ As a simple example, consider the following:
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp .. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++ :language: c++
:lines: 111-124 :lines: 137-151
:emphasize-lines: 8-10,13 :emphasize-lines: 8-12,14
:linenos: :linenos:
2. Configure cmake with **WITH_TIMER** and recompile PaddlePaddle. 2. Configure cmake with **WITH_TIMER** and recompile PaddlePaddle.
...@@ -90,31 +90,31 @@ As a simple example, consider the following: ...@@ -90,31 +90,31 @@ As a simple example, consider the following:
cmake .. -DWITH_TIMER=ON cmake .. -DWITH_TIMER=ON
make make
3. Execute your code and observe the results (see the emphasize-lines). 3. Execute your code and observe the results (see the emphasize-lines).
.. code-block:: bash .. code-block:: bash
:emphasize-lines: 1,12-15 :emphasize-lines: 1,12-15
> ./paddle/math/tests/test_GpuProfiler > ./paddle/math/tests/test_GpuProfiler
I1117 11:13:42.313065 2522362816 Util.cpp:155] commandline: ./paddle/math/tests/test_GpuProfiler I1117 11:13:42.313065 2522362816 Util.cpp:155] commandline: ./paddle/math/tests/test_GpuProfiler
I1117 11:13:42.845065 2522362816 Util.cpp:130] Calling runInitFunctions I1117 11:13:42.845065 2522362816 Util.cpp:130] Calling runInitFunctions
I1117 11:13:42.845208 2522362816 Util.cpp:143] Call runInitFunctions done. I1117 11:13:42.845208 2522362816 Util.cpp:143] Call runInitFunctions done.
[==========] Running 1 test from 1 test case. [==========] Running 1 test from 1 test case.
[----------] Global test environment set-up. [----------] Global test environment set-up.
[----------] 1 test from Profiler [----------] 1 test from Profiler
[ RUN ] Profiler.BilinearFwdBwd [ RUN ] Profiler.BilinearFwdBwd
I1117 11:13:42.845310 2522362816 test_GpuProfiler.cpp:114] Enable GPU Profiler Stat: [testBilinearFwdBwd] "numSamples = 10, channels = 16, im I1117 11:13:42.845310 2522362816 test_GpuProfiler.cpp:114] Enable GPU Profiler Stat: [testBilinearFwdBwd] "numSamples = 10, channels = 16, im
gSizeX = 64, imgSizeY = 64" gSizeX = 64, imgSizeY = 64"
I1117 11:13:42.850154 2522362816 ThreadLocal.cpp:37] thread use undeterministic rand seed:20659751 I1117 11:13:42.850154 2522362816 ThreadLocal.cpp:37] thread use undeterministic rand seed:20659751
I1117 11:13:42.981501 2522362816 Stat.cpp:130] ======= StatSet: [GlobalStatInfo] status ====== I1117 11:13:42.981501 2522362816 Stat.cpp:130] ======= StatSet: [GlobalStatInfo] status ======
I1117 11:13:42.981539 2522362816 Stat.cpp:133] Stat=testBilinearFwdBwd total=136.141 avg=136.141 max=136.141 min=136.141 count=1 I1117 11:13:42.981539 2522362816 Stat.cpp:133] Stat=testBilinearFwdBwd total=136.141 avg=136.141 max=136.141 min=136.141 count=1
I1117 11:13:42.981572 2522362816 Stat.cpp:141] ======= BarrierStatSet status ====== I1117 11:13:42.981572 2522362816 Stat.cpp:141] ======= BarrierStatSet status ======
I1117 11:13:42.981575 2522362816 Stat.cpp:154] -------------------------------------------------- I1117 11:13:42.981575 2522362816 Stat.cpp:154] --------------------------------------------------
[ OK ] Profiler.BilinearFwdBwd (136 ms) [ OK ] Profiler.BilinearFwdBwd (136 ms)
[----------] 1 test from Profiler (136 ms total) [----------] 1 test from Profiler (136 ms total)
[----------] Global test environment tear-down [----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (136 ms total) [==========] 1 test from 1 test case ran. (136 ms total)
[ PASSED ] 1 test. [ PASSED ] 1 test.
nvprof profiler nvprof profiler
...@@ -126,7 +126,7 @@ To use this command line profiler **nvprof**, you can simply issue the following ...@@ -126,7 +126,7 @@ To use this command line profiler **nvprof**, you can simply issue the following
.. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp .. literalinclude:: ../../../paddle/math/tests/test_GpuProfiler.cpp
:language: c++ :language: c++
:lines: 111-124 :lines: 137-151
:emphasize-lines: 6-7 :emphasize-lines: 6-7
:linenos: :linenos:
...@@ -147,42 +147,42 @@ Then, you can get the following profiling result: ...@@ -147,42 +147,42 @@ Then, you can get the following profiling result:
.. code-block:: bash .. code-block:: bash
==78544== Profiling application: ./paddle/math/tests/test_GpuProfiler ==78544== Profiling application: ./paddle/math/tests/test_GpuProfiler
==78544== Profiling result: ==78544== Profiling result:
Time(%) Time Calls Avg Min Max Name Time(%) Time Calls Avg Min Max Name
27.60% 9.6305ms 5 1.9261ms 3.4560us 6.4035ms [CUDA memcpy HtoD] 27.60% 9.6305ms 5 1.9261ms 3.4560us 6.4035ms [CUDA memcpy HtoD]
26.07% 9.0957ms 1 9.0957ms 9.0957ms 9.0957ms KeBilinearInterpBw 26.07% 9.0957ms 1 9.0957ms 9.0957ms 9.0957ms KeBilinearInterpBw
23.78% 8.2977ms 1 8.2977ms 8.2977ms 8.2977ms KeBilinearInterpFw 23.78% 8.2977ms 1 8.2977ms 8.2977ms 8.2977ms KeBilinearInterpFw
22.55% 7.8661ms 2 3.9330ms 1.5798ms 6.2863ms [CUDA memcpy DtoH] 22.55% 7.8661ms 2 3.9330ms 1.5798ms 6.2863ms [CUDA memcpy DtoH]
==78544== API calls: ==78544== API calls:
Time(%) Time Calls Avg Min Max Name Time(%) Time Calls Avg Min Max Name
46.85% 682.28ms 8 85.285ms 12.639us 682.03ms cudaStreamCreateWithFlags 46.85% 682.28ms 8 85.285ms 12.639us 682.03ms cudaStreamCreateWithFlags
39.83% 580.00ms 4 145.00ms 302ns 550.27ms cudaFree 39.83% 580.00ms 4 145.00ms 302ns 550.27ms cudaFree
9.82% 143.03ms 9 15.892ms 8.7090us 142.78ms cudaStreamCreate 9.82% 143.03ms 9 15.892ms 8.7090us 142.78ms cudaStreamCreate
1.23% 17.983ms 7 2.5690ms 23.210us 6.4563ms cudaMemcpy 1.23% 17.983ms 7 2.5690ms 23.210us 6.4563ms cudaMemcpy
1.23% 17.849ms 2 8.9247ms 8.4726ms 9.3768ms cudaStreamSynchronize 1.23% 17.849ms 2 8.9247ms 8.4726ms 9.3768ms cudaStreamSynchronize
0.66% 9.5969ms 7 1.3710ms 288.43us 2.4279ms cudaHostAlloc 0.66% 9.5969ms 7 1.3710ms 288.43us 2.4279ms cudaHostAlloc
0.13% 1.9530ms 11 177.54us 7.6810us 591.06us cudaMalloc 0.13% 1.9530ms 11 177.54us 7.6810us 591.06us cudaMalloc
0.07% 1.0424ms 8 130.30us 1.6970us 453.72us cudaGetDevice 0.07% 1.0424ms 8 130.30us 1.6970us 453.72us cudaGetDevice
0.04% 527.90us 40 13.197us 525ns 253.99us cudaEventCreateWithFlags 0.04% 527.90us 40 13.197us 525ns 253.99us cudaEventCreateWithFlags
0.03% 435.73us 348 1.2520us 124ns 42.704us cuDeviceGetAttribute 0.03% 435.73us 348 1.2520us 124ns 42.704us cuDeviceGetAttribute
0.03% 419.36us 1 419.36us 419.36us 419.36us cudaGetDeviceCount 0.03% 419.36us 1 419.36us 419.36us 419.36us cudaGetDeviceCount
0.02% 260.75us 2 130.38us 129.32us 131.43us cudaGetDeviceProperties 0.02% 260.75us 2 130.38us 129.32us 131.43us cudaGetDeviceProperties
0.02% 222.32us 2 111.16us 106.94us 115.39us cudaLaunch 0.02% 222.32us 2 111.16us 106.94us 115.39us cudaLaunch
0.01% 214.06us 4 53.514us 28.586us 77.655us cuDeviceGetName 0.01% 214.06us 4 53.514us 28.586us 77.655us cuDeviceGetName
0.01% 115.45us 4 28.861us 9.8250us 44.526us cuDeviceTotalMem 0.01% 115.45us 4 28.861us 9.8250us 44.526us cuDeviceTotalMem
0.01% 83.988us 4 20.997us 578ns 77.760us cudaSetDevice 0.01% 83.988us 4 20.997us 578ns 77.760us cudaSetDevice
0.00% 38.918us 1 38.918us 38.918us 38.918us cudaEventCreate 0.00% 38.918us 1 38.918us 38.918us 38.918us cudaEventCreate
0.00% 34.573us 31 1.1150us 279ns 12.784us cudaDeviceGetAttribute 0.00% 34.573us 31 1.1150us 279ns 12.784us cudaDeviceGetAttribute
0.00% 17.767us 1 17.767us 17.767us 17.767us cudaProfilerStart 0.00% 17.767us 1 17.767us 17.767us 17.767us cudaProfilerStart
0.00% 15.228us 2 7.6140us 3.5460us 11.682us cudaConfigureCall 0.00% 15.228us 2 7.6140us 3.5460us 11.682us cudaConfigureCall
0.00% 14.536us 2 7.2680us 1.1490us 13.387us cudaGetLastError 0.00% 14.536us 2 7.2680us 1.1490us 13.387us cudaGetLastError
0.00% 8.6080us 26 331ns 173ns 783ns cudaSetupArgument 0.00% 8.6080us 26 331ns 173ns 783ns cudaSetupArgument
0.00% 5.5470us 6 924ns 215ns 2.6780us cuDeviceGet 0.00% 5.5470us 6 924ns 215ns 2.6780us cuDeviceGet
0.00% 5.4090us 6 901ns 328ns 3.3320us cuDeviceGetCount 0.00% 5.4090us 6 901ns 328ns 3.3320us cuDeviceGetCount
0.00% 4.1770us 3 1.3920us 1.0630us 1.8300us cuDriverGetVersion 0.00% 4.1770us 3 1.3920us 1.0630us 1.8300us cuDriverGetVersion
0.00% 3.4650us 3 1.1550us 1.0810us 1.2680us cuInit 0.00% 3.4650us 3 1.1550us 1.0810us 1.2680us cuInit
0.00% 830ns 1 830ns 830ns 830ns cudaRuntimeGetVersion 0.00% 830ns 1 830ns 830ns 830ns cudaRuntimeGetVersion
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或