# Select Denver core specifically for execution

**URL:** <https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654>\
**Category:** Jetson TX2\
**Created:** [September 23, 2017, 1:13am UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654 "2017-09-23T01:13:22Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![xtracrispy](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@xtracrispy](https://forums.developer.nvidia.com/u/xtracrispy)\
**Post date:** [September 23, 2017, 1:13am UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654/1 "2017-09-23T01:13:22Z")

</div>

Hello,

Is there a way to select which core a thread runs on?  
Previously, I was getting good performance when Denver was automatically selected for a program I’m writing.  
Recently, I ported some of the processing to GPU, which reduced demands greatly.

It seems that I’m now on one of the slower cores, and it’s pegged to 100% for some net performance degradation. Is there a way to have the best of both worlds, and run the host application on Denver?

---

<div class="post-metadata">

**Author:** ![dusty\_nv](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/dusty_nv/32/14043_2.png) [@dusty\_nv](https://forums.developer.nvidia.com/u/dusty_nv)\
**Post date:** [September 23, 2017, 2:00pm UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654/2 "2017-09-23T14:00:13Z")

</div>

To park the CPU affinity, I think you are looking for the [taskset](https://linux.die.net/man/1/taskset) utility (which parks a process) or the [pthread\_setaffinity](https://computing.llnl.gov/tutorials/pthreads/man/pthread_setaffinity_np.txt) API (to park individual threads).

Note that upon boot-up of TX2, the Denver cores are disabled to save power. Enable them with [nvpmodel](http://www.jetsonhacks.com/2017/03/25/nvpmodel-nvidia-jetson-tx2-development-kit/) tool and verify by running tegrastats.

---

<div class="post-metadata">

**Author:** ![ericbin](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@ericbin](https://forums.developer.nvidia.com/u/ericbin)\
**Post date:** [September 27, 2017, 3:32am UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654/3 "2017-09-27T03:32:10Z")

</div>

mark

---

<div class="post-metadata">

**Author:** ![disheet](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@disheet](https://forums.developer.nvidia.com/u/disheet)\
**Post date:** [April 13, 2018, 5:04am UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654/4 "2018-04-13T05:04:37Z")

</div>

Hello xtracrispy,

Can you tell me how you ported some of the processing to GPU ?

---

<div class="post-metadata">

**Author:** ![nadeemm](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/nadeemm/32/367020_2.png) [@nadeemm](https://forums.developer.nvidia.com/u/nadeemm)\
**Post date:** [October 18, 2021, 6:28pm UTC](https://forums.developer.nvidia.com/t/select-denver-core-specifically-for-execution/53654/5 "2021-10-18T18:28:42Z")

</div>