# How to use the LDS.128 

**URL:** <https://forums.developer.nvidia.com/t/how-to-use-the-lds-128/27777>\
**Category:** CUDA Programming and Performance\
**Created:** [December 27, 2012, 8:19am UTC](https://forums.developer.nvidia.com/t/how-to-use-the-lds-128/27777 "2012-12-27T08:19:28Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![cuda123](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@cuda123](https://forums.developer.nvidia.com/u/cuda123)\
**Post date:** [December 27, 2012, 8:19am UTC](https://forums.developer.nvidia.com/t/how-to-use-the-lds-128/27777/1 "2012-12-27T08:19:28Z")

</div>

I had seen that there has an instruction of LDS.128. But is anybody known how to writing a C code to generate this instruction? thanks!

---

<div class="post-metadata">

**Author:** ![njuffa](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@njuffa](https://forums.developer.nvidia.com/u/njuffa)\
**Post date:** [December 27, 2012, 9:46pm UTC](https://forums.developer.nvidia.com/t/how-to-use-the-lds-128/27777/2 "2012-12-27T21:46:26Z")

</div>

If your code loads double2, float4, or int4 data from shared memory, you should see a LDS.128 being generated (provided your code uses all components of the vector, otherwise the compiler may optimize the code to use narrower loads).