# Graphical interface for cuda-gdb?

**URL:** <https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458>\
**Category:** CUDA Programming and Performance\
**Created:** [January 28, 2009, 5:56pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458 "2009-01-28T17:56:06Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![waswas](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@waswas](https://forums.developer.nvidia.com/u/waswas)\
**Post date:** [January 28, 2009, 5:56pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/1 "2009-01-28T17:56:06Z")

</div>

Is there a ddd (or GUI) available for cuda-gdb?

---

<div class="post-metadata">

**Author:** ![tmurray](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@tmurray](https://forums.developer.nvidia.com/u/tmurray)\
**Post date:** [January 28, 2009, 7:34pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/2 "2009-01-28T19:34:19Z")

</div>

Supposedly ddd just works with it, but I’ve never used it. Let me ask around…

---

<div class="post-metadata">

**Author:** ![waswas](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@waswas](https://forums.developer.nvidia.com/u/waswas)\
**Post date:** [January 28, 2009, 7:55pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/3 "2009-01-28T19:55:35Z")

</div>

How do you configure the ddd installation to integrate with cuda-gdb instead of host-only /usr/bin/gdb?

---

<div class="post-metadata">

**Author:** ![tmurray](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@tmurray](https://forums.developer.nvidia.com/u/tmurray)\
**Post date:** [January 28, 2009, 7:56pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/4 "2009-01-28T19:56:21Z")

</div>

ddd --debugger cuda-gdb says the debugger team. I think they were mildly annoyed at me for asking instead of just doing man ddd myself :(

---

<div class="post-metadata">

**Author:** ![waswas](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@waswas](https://forums.developer.nvidia.com/u/waswas)\
**Post date:** [January 29, 2009, 12:24pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/5 "2009-01-29T12:24:20Z")

</div>

Thank you for finding the answer to my question.

I didn’t see any reference to a GUI in the cuda-gdb User Manual, so one never knows whether there are special configurations beyond just switching the console debugger…

---

<div class="post-metadata">

**Author:** ![Thorn\_Striff](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/thorn_striff/32/23243_2.png) [@Thorn\_Striff](https://forums.developer.nvidia.com/u/Thorn_Striff)\
**Post date:** [May 29, 2010, 6:21pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/6 "2010-05-29T18:21:21Z")

</div>

Isnt there any way to run DDD with cuda-gdb without need 2 graphic cards?

---

<div class="post-metadata">

**Author:** ![Thorn\_Striff](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/thorn_striff/32/23243_2.png) [@Thorn\_Striff](https://forums.developer.nvidia.com/u/Thorn_Striff)\
**Post date:** [May 29, 2010, 6:30pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/7 "2010-05-29T18:30:19Z")

</div>

Isnt there any way to run DDD with cuda-gdb without need 2 graphic cards?

---

<div class="post-metadata">

**Author:** ![mcleary](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/mcleary/32/28470_2.png) [@mcleary](https://forums.developer.nvidia.com/u/mcleary)\
**Post date:** [January 14, 2011, 5:09pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/8 "2011-01-14T17:09:51Z")

</div>

> [@](#):
>
> Isnt there any way to run DDD with cuda-gdb without need 2 graphic cards?

The _GETTING STARTED GUIDE FOR cuda-gdb PREVIEW RELEASE_ says the following

> [@](#):
>
> When the graphics desktop manager is running on the same GPU as CUDA then
> 
> debugging that GPU results in pausing of the GPU which also pauses the desktop and
> 
> thus freezes the GUI which makes the desktop unusable. To avoid desktop hangs
> 
> cuda-gdb can be used in the following system configurations.

So it’s not possible to use a GUI with cuda-gdb e a single card system.

---

<div class="post-metadata">

**Author:** ![laughingrice](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@laughingrice](https://forums.developer.nvidia.com/u/laughingrice)\
**Post date:** [January 16, 2011, 7:52pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/9 "2011-01-16T19:52:17Z")

</div>

> [@](#):
>
> The _GETTING STARTED GUIDE FOR cuda-gdb PREVIEW RELEASE_ says the following
> 
> So it’s not possible to use a GUI with cuda-gdb e a single card system.

You can do it with two machines via ssh -X. You may also find emacs useful as it has both a gdb mode and can run in the console. Not as gui as ddd, but possibly good enough.

And you can console yourself with the fact that windows doesn’t have a single card / debug in console option.

---

<div class="post-metadata">

**Author:** ![mcleary](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/mcleary/32/28470_2.png) [@mcleary](https://forums.developer.nvidia.com/u/mcleary)\
**Post date:** [January 16, 2011, 10:11pm UTC](https://forums.developer.nvidia.com/t/graphical-interface-for-cuda-gdb/7458/10 "2011-01-16T22:11:40Z")

</div>

> [@](#):
>
> You can do it with two machines via ssh -X. You may also find emacs useful as it has both a gdb mode and can run in the console. Not as gui as ddd, but possibly good enough.
> 
> And you can console yourself with the fact that windows doesn’t have a single card / debug in console option.

The problem with ssh -X is that is not possible to debug a graphical application. Since VBO’s resides in GPU memory, there’s no way ssh can send these data through a network, at least I wasn’t able to do this.
