Skip to content

Commit

Permalink
Deployed 99e4a91 with MkDocs version: 1.5.3
Browse files Browse the repository at this point in the history
  • Loading branch information
Unknown committed Jan 31, 2024
1 parent 6423adb commit cf9ddc4
Show file tree
Hide file tree
Showing 3 changed files with 11 additions and 10 deletions.
2 changes: 1 addition & 1 deletion search/search_index.json

Large diffs are not rendered by default.

Binary file modified sitemap.xml.gz
Binary file not shown.
19 changes: 10 additions & 9 deletions user-guide/gpu/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -1487,7 +1487,7 @@
<li class="md-nav__item">
<a href="#hipify" class="md-nav__link">
<span class="md-ellipsis">
Hipify
HIPIFY
</span>
</a>

Expand Down Expand Up @@ -3456,7 +3456,7 @@
<li class="md-nav__item">
<a href="#hipify" class="md-nav__link">
<span class="md-ellipsis">
Hipify
HIPIFY
</span>
</a>

Expand Down Expand Up @@ -3578,8 +3578,9 @@ <h3 id="programming-environments">Programming Environments</h3>
</tbody>
</table>
<p>To decide which compiler(s) to use to compile offload code for the AMD
GPUs, you may find it useful to consult the <a href="#compilation-strategies">Compilation
Strategies</a> section below.</p>
GPUs, you may find it useful to consult the <a href="#compilation-strategies-for-gpu-offloading">Compilation Strategies
for GPU Offloading</a>
section below.</p>
<p>The hybrid environments <code>PrgEnv-gnu-amd</code> and <code>PrgEnv-cray-amd</code> are
provided as a convenient way to mitigate less mature OpenMP offload
support in the AMD LLVM Fortran compiler. In these hybrid environments
Expand Down Expand Up @@ -4485,11 +4486,11 @@ <h3 id="rocm-smi">rocm-smi</h3>
<p><code>rocm-smi --help</code></p>
<p>will run on the login nodes to get more infomation about probing the GPUs.</p>
<p>More detail can be found at <a href="https://github.com/ROCm/rocm_smi_lib/tree/rocm-5.2.3/python_smi_tools">here</a>.</p>
<h3 id="hipify">Hipify</h3>
<p>Is a CUDA to HIP source translator tool that can allow CUDA source code to be translated into HIP source code easing the transition between the two hardware targets.</p>
<p>The tool is avalible on ARCHER2 by loading the <code>rocm</code> module.</p>
<p>The github repository for HIPify can be found <a href="https://github.com/ROCm/HIPIFY">here</a>.</p>
<p>The documentation for hipfiy is found <a href="https://rocm.docs.amd.com/projects/HIPIFY/en/docs-5.2.3/index.html">here</a>.</p>
<h3 id="hipify">HIPIFY</h3>
<p>HIPIFY is a CUDA to HIP source translator tool that can allow CUDA source code to be translated into HIP source code easing the transition between the two hardware targets.</p>
<p>The tool is available on ARCHER2 by loading the <code>rocm</code> module.</p>
<p>The github repository for HIPIFY can be found <a href="https://github.com/ROCm/HIPIFY">here</a>.</p>
<p>The documentation for HIPIFY is found <a href="https://rocm.docs.amd.com/projects/HIPIFY/en/docs-5.2.3/index.html">here</a>.</p>
<h2 id="notes-and-useful-links">Notes and useful links</h2>
<p>You should expect the software development environment to be similar to that
available on the Frontier exascale system:</p>
Expand Down

0 comments on commit cf9ddc4

Please sign in to comment.