CUDA

Open topic with navigation

You are here: How To: Set Breakpoints in a Shader > Walkthrough: Debugging a CUDA Application > CUDA Project Properties

CUDA Project Properties

NVIDIA® Nsight™ Application Development Environment for Heterogeneous Platforms, Visual Studio Edition 6.0 User Guide 
Send Feedback 

---

With NVIDIA Nsight, you can set parameters of your CUDA project in order to customize your debugging experience.

To configure your project's CUDA properties page:

1. In the Solution Explorer, click on the project name so that it is highlighted.
2. From the Project menu, choose Properties. The Property Pages window opens.
3. Select CUDA C/C++ in the left pane.

Common

On the Common page, you can configure the following options:

1. CUDA Toolkit Custom Dir — This option sets a custom path to the CUDA toolkit. You can edit the path, select "Browse" to choose the path, or select "inherit from parent or project defaults." 
2. Source Dependencies — This option allows you to add additional source file dependencies. If you have a dependency that has been set with an #include statement, it does not need to be explicitly specified in this
3. Compiler Output (obj/cubin) — This sets the output as a an .obj or a .cubin file. The default setting is $(IntDir)%(Filename)%(Extension).obj.
4. Additional Include Directories — This option allows you to list at least one additional directory to add to the include path. If you have more than one, use a semi-colon to separate them.
5. Use Host Include Directories — This option specifies whether or not to use the additional include directories that are used by the host compiler for device code.
6. Keep Preprocessed Files — This option allows you to choose whether or not the preprocessor files generated by the CUDA compiler (for example, .ptx, .cubin, .cudafe1.c, etc.) will be deleted.
7. Keep Directory — This option sets the path the directory where the preprocessor files generated by the CUDA compiler will be kept.
8. Generate Relocatable Device Code — This setting chooses whether or not to compile the input file into an object file that contains relocatable device code.
9. NVCC Compilation Type — This option sets your desired output of NVCC compilation. Choices here include the following: 
   • Generate hybrid object file (--compile) 
   • Generate hybrid .c file (-cuda) 
   • Generate .gpu file (-gpu) 
   • Generate .cubin file (-cubin) 
   • Generate .ptx file (-ptx) 
10. CUDA Runtime — This option allows you to specify the type of CUDA runtime library to be used. The choices here include the following: 
    • No CUDA runtime library (-cudart none) 
    • Shared/dynamic CUDA runtime library (-cudart shared) 
    • Static CUDA runtime library (-cudart static) 
11. Target Machine Platform — This sets the platform of the target machine (either x86 or x64).

Device

On the Device page, you can configure the following options: 

1. C interleaved in PTXAS Output — This setting chooses whether or not to insert source code into generated PTX.
2. Code Generation — This option specifies the names of the NVIDIA GPU architectures to generate code for. If you click Edit from the drop-down menu, the following pop-up appears: 

   

   If you edit this field, the correct syntax to use is [arch],[code] (for example, compute_10,sm_10). If the selected NVCC Compilation Type is compile, then multiple arch/code pairs may be listed, separated by a semi-colon (for example, compute_10,sm_10;compute_20,sm_20).

   Valid values for arch and code include the following: 

   • compute_10
   • compute_11
   • compute_12
   • compute_13
   • compute_20
   • compute_30
   • sm_10
   • sm_11
   • sm_12
   • sm_13
   • sm_20
   • sm_21
   • sm_30
   • sm_35 
3. Generate GPU Debug Information — This setting selects whether or not GPU debugging information is generated by the CUDA compiler.
4. Generate Line Number Information — This option chooses whether or not to generate line number information for device code. If Generate GPU Debug Information is on (-G), line information (-lineinfo) is automatically generated as well.
5. Max Used Register — This option specifies the maximum amount of registers that GPU functions can use.
6. Verbose PTXAS Output — This option selects whether or not to use verbose PTXAS output.

Host 

On the Host page, you can configure the following options: 

1. Additional Compiler Options — This setting lists additional host compiler options that are not supported by the host's project properties.
2. Preprocessor Definitions — This option allows you to list preprocessor defines.
3. Use Host Preprocessor Definitions — This option selects whether or not to use the defines that were used by the host compiler for device code.
4. Emulation — This option specifies whether or not to generate emulated code.
5. Generate Host Debug Information — This option specifies whether or not the host debugging information will be generated by the CUDA compiler.
6. Use Fast Math — This option selects whether or not to make use of the fast math library.
7. Optimization — This field selects the option for code optimization. Available choices include the following: 
   • <inherit from host> 
   • Disabled (/Od)
   • Minimize Size (/O1) 
   • Maximize Speed (/O2) 
   • Full Optimization (/Ox)
8. Runtime Library — This field selects the runtime library to use for linking. Available choices include the following: 
   • <inherit from host> 
   • Multi-Threaded (/Mt)
   • Multi-Threaded Debug (/Mtd) 
   • Multi-Threaded DLL (/MD)
   • Multi-Threaded Debug DLL (/MDd)
   • Single-Threaded (/ML) 
   • Single-Threaded Debug (/MLd)
9. Basic Runtime Checks — This field performs basic runtime error checks, incompatible with any optimization type other than debug. Available choices include the following: 
   • <inherit from host>
   • Default
   • Stack Frames (/RTCs)
   • Uninitialized Variables (/RTCu)
   • Both (/RTC1) 
10. Enable Run-Time Type Info — This option chooses whether or not to add code for checking C++ object types at run time.
11. Warning Level — This option selects how strictly you want the compiler to be when checking for potentially suspect constructs. Available choices here include: 
    • <inherit from host> 
    • Off: Turn Off All Warnings (/W0)
    • Level 1 (/W1)
    • Level 2 (/W2)
    • Level 3 (/W3)
    • Level 4 (/W4)
    • Enable All Warnings (/Wall)

Command Line

The Command Line page shows the approximate command line parameters, given the settings you've chosen.

 

 

 

---

NVIDIA® Nsight™ Application Development Environment for Heterogeneous Platforms, Visual Studio Edition User Guide Rev. 6.0.180917 ©2009-2018. NVIDIA Corporation. All Rights Reserved.

Open topic with navigation