<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto">Hi,<div><br></div><div>Without going into too much details (which I can't :-), there have been some efforts that were not too happy with hip performance. </div><div><br></div><div>We haven't yet had time to look into it ourselves, but it'son our radar, and with at least one concrete hip port example (say, for one gromacs kernel) where it's appreciably faster than OpenCL it would likely get higher priority in the team.</div><div><br></div><div>So, the short story is that it's not supported in release 2020, but it might appear in master during the year.</div><div><br></div><div>Cheers,</div><div><br></div><div>Erik</div><div><br></div><div><br></div><div><br><div dir="ltr">--<div>Erik Lindahl <erik.lindahl@scilifelab.se></div><div>Professor of Biophysics</div><div>Science for Life Laboratory</div><div>Stockholm University & KTH</div><div>Office (SciLifeLab): +46 8 524 81567</div><div>Cell (Sweden): +46 73 4618050 </div><div>Cell (US): 1 267 307 8746</div><div><br></div></div><div dir="ltr"><br><blockquote type="cite">On Jan 4, 2020, at 08:53, Benson Muite <benson_muite@emailplus.org> wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<p>Hi Zhang,</p>
<p>This is interesting, though maintaining a large number of
accelerated versions may be challenging. OpenCL is more portable
than HIP, Vulkan looks like a possible well supported accelerator
API. Is your code openly available?</p>
<p>Benson<br>
</p>
<div class="moz-cite-prefix">On 1/4/20 9:49 AM, 张驭洲 wrote:<br>
</div>
<blockquote type="cite" cite="mid:4cc7bba1.1db1e.16f6f50862b.Coremail.zhangyuzhou15@mails.ucas.edu.cn">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<p> Dear gromacs developers,
</p>
<p> I want to know if there is any HIP version of gromacs.HIP is a
thin layer above the AMD Rocm stack, and it allows CUDA code to
be converted to portable C++. You can find more details about
HIP on <a href="https://github.com/ROCm-Developer-Tools/HIP" moz-do-not-send="true">https://github.com/ROCm-Developer-Tools/HIP</a>. Currently
gromacs runs on AMD GPU via OpenCL,but the performance of the
OpenCL version may be worse than the CUDA version.
</p>
<p> Actually I have a HIP version of gromacs right now, which is
converted from 2018.7. I have tested it on AMD MI50 GPUs and get
10%~30% better performance on some test cases than the OpenCL
code of gromacs 2019.3. I want to know if there is any official
plan to develop HIP version gromacs, or anyone else who is
interested in this work.
</p>
<p> </p>
<p> </p>
<p> <span style="text-align:justify;color:#333333;text-transform:none;text-indent:0px;letter-spacing:normal;font-family:arial;font-size:16px;font-style:normal;font-weight:400;word-spacing:0px;float:none;display:inline
!important;white-space:normal;orphans:2;widows:2;background-color:#FFFFFF;-webkit-text-stroke-width:0px;font-variant-ligatures:normal;font-variant-caps:normal;text-decoration-style:initial;text-decoration-color:initial;"><br>
</span> </p>
Sincerely,
<p> zhang </p>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
</blockquote>
<span>-- </span><br><span>Gromacs Developers mailing list</span><br><span></span><br><span>* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!</span><br><span></span><br><span>* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists</span><br><span></span><br><span>* For (un)subscribe requests visit</span><br><span>https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request@gromacs.org.</span></div></blockquote></div></body></html>