<html><head></head><body><div class="ydp9fdc2eafyahoo-style-wrap" style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:16px;"><div><div>OK I found that you are referring to ....kernels.cuh while I was looking at ....kernel.cuh. There, I see</div><div><br></div><div><br></div><div><span>/* cut-off + V shift LJ w LB combination rules */<br>#define LJ_COMB_LB<br>#define NB_KERNEL_FUNC_NAME(x, ...) x ## _ElecEw_VdwLJCombLB ## __VA_ARGS__<br>#include FLAVOR_LEVEL_GENERATOR<br></span><div><span><br></span></div><div><br></div><div><br></div><div>That is the kernel name and how that is run in <span>nbnxn_cuda_kernel.cuh</span> is what I described in the previous post and you confirmed that.</div><div><br></div><div>Thank you very much. <br></div><div><br><span></span></div></div><div><br></div><div class="ydp9fdc2eafsignature">Regards,<br>Mahmood</div></div>
<div><br></div><div><br></div>
</div><div id="yahoo_quoted_6820374609" class="yahoo_quoted">
<div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
<div>
On Thursday, April 25, 2019, 5:21:50 PM GMT+4:30, Berk Hess <hess@kth.se> wrote:
</div>
<div><br></div>
<div><br></div>
<div><div id="yiv0654810570"><div>
<div class="yiv0654810570moz-cite-prefix">Also, you are grepping for LJ_Comb, not
LJComb as Szilard suggested.<br clear="none">
<br clear="none">
Cheers,<br clear="none">
<br clear="none">
Berk<br clear="none">
<br clear="none">
On 25/04/2019 14.48, Szilárd Páll wrote:<br clear="none">
</div>
<blockquote type="cite">
</blockquote></div><div class="yiv0654810570yqt4931552956" id="yiv0654810570yqt42122"><div><div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr"><br clear="all">
<div>
<div class="yiv0654810570gmail_signature" dir="ltr"><br clear="none">
</div>
</div>
<br clear="none">
</div>
<br clear="none">
<div class="yiv0654810570gmail_quote">
<div class="yiv0654810570gmail_attr" dir="ltr">On Thu, Apr 25, 2019
at 2:43 PM Mahmood Naderan <<a rel="nofollow" shape="rect" ymailto="mailto:nt_mahmood@yahoo.com" target="_blank" href="mailto:nt_mahmood@yahoo.com">nt_mahmood@yahoo.com</a>>
wrote:<br clear="none">
</div>
<blockquote class="yiv0654810570gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex;">
<div>
<div class="yiv0654810570gmail-m_-7076311669053638246ydp470688b9yahoo-style-wrap" style="">
<div>
<div><span>> <span>Selection happens at
runtime, see
in nbnxn_cuda.cu:select_nbnxn_kernel().</span></span></div>
<div><span><span><br clear="none">
</span></span></div>
<div><span><span>Yes I see that.</span></span></div>
<div><span><span><br clear="none">
</span></span>
<div><br clear="none">
</div>
<span><span></span></span></div>
<div><span><br clear="none">
</span></div>
<div><span><br clear="none">
</span></div>
<div><span>> Try "git grep LJComb".</span></div>
<div><span><br clear="none">
</span></div>
<div><span>I am trying to find that in 2018.2
release that I have currently.</span></div>
<div><span><br clear="none">
</span></div>
<div><span><span><a rel="nofollow" shape="rect" class="yiv0654810570moz-txt-link-abbreviated" ymailto="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$" target="_blank" href="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$">mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$</a>
grep -i LJ_Comb nbnxn_cuda_kernel.cuh<br clear="none">
#if defined LJ_COMB_GEOM || defined
LJ_COMB_LB<br clear="none">
#define LJ_COMB<br clear="none">
#ifndef LJ_COMB<br clear="none">
const float2 *lj_comb =
atdat.lj_comb;<br clear="none">
#ifndef LJ_COMB<br clear="none">
#if !defined LJ_COMB_LB || defined
CALC_ENERGIES<br clear="none">
#ifdef LJ_COMB_LB<br clear="none">
#ifndef LJ_COMB<br clear="none">
#ifndef LJ_COMB<br clear="none">
ljcpib[tidxj * c_clSize + tidxi] =
lj_comb[ai];<br clear="none">
#ifndef LJ_COMB<br clear="none">
ljcp_j = lj_comb[aj];<br clear="none">
#ifndef LJ_COMB<br clear="none">
#ifdef LJ_COMB_GEOM<br clear="none">
#endif /*
LJ_COMB_GEOM */<br clear="none">
#endif /* LJ_COMB
*/<br clear="none">
#if !defined LJ_COMB_LB || defined
CALC_ENERGIES<br clear="none">
#else /*
!LJ_COMB_LB || CALC_ENERGIES */<br clear="none">
#endif /*
!LJ_COMB_LB || CALC_ENERGIES */<br clear="none">
#undef LJ_COMB<br clear="none">
<a rel="nofollow" shape="rect" class="yiv0654810570moz-txt-link-abbreviated" ymailto="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$" target="_blank" href="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$">mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$</a> grep -i
LJComb nbnxn_cuda_kernel.cuh<br clear="none">
</span></span>
<div><span><span><a rel="nofollow" shape="rect" class="yiv0654810570moz-txt-link-abbreviated" ymailto="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$" target="_blank" href="mailto:mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$">mahmood@mpc:/opt/gromacs-2018.2/src/gromacs/mdlib/nbnxn_cuda$</a></span></span></div>
</div>
</div>
</div>
</div>
</blockquote>
<div><br clear="none">
</div>
<div>As suggested before, grep the source repo, not that
one file; or grep the right file -- the one I
recommended looking at in my previous email
(src/gromacs/mdlib/nbnxn_cuda/nbnxn_cuda_kernels.cuh).</div>
<div> </div>
<blockquote class="yiv0654810570gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex;">
<div>
<div class="yiv0654810570gmail-m_-7076311669053638246ydp470688b9yahoo-style-wrap" style="">
<div>
<div>
<div><span><span></span></span><br clear="none">
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div>I expect that <span><span><span>VdwLJCombLB_F</span></span></span>
means that the statements that must be
executed in <span>nbnxn_cuda_kernel.cuh,
are those that are used in defined macros.</span></div>
</div>
</div>
</div>
</div>
</blockquote>
<div><br clear="none">
</div>
<div>Correct. In fact the "VdwLJCombLB" kenrnels will
not even have e.g. VdwLJFsw (i.e. LJ force switch)
code included in them.</div>
<div> </div>
<blockquote class="yiv0654810570gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex;">
<div>
<div class="yiv0654810570gmail-m_-7076311669053638246ydp470688b9yahoo-style-wrap" style="">
<div>
<div>
<div><span><br clear="none">
</span></div>
<div><span>For example, since I see Vdw, the
following code must be executed</span></div>
<div><span><br clear="none">
</span></div>
<div><span><span>#ifdef VDW_CUTOFF_CHECK<br clear="none">
/*
Separate VDW cut-off check to enable
twin-range cut-offs<br clear="none">
* (rvdw
< rcoulomb <= rlist)<br clear="none">
*/<br clear="none">
vdw_in_range = (r2 < rvdw_sq) ? 1.0f
: 0.0f;<br clear="none">
F_invr *= vdw_in_range;<br clear="none">
<br clear="none">
</span></span>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div>Also, since LJCombLB is seen, the
following code must be executed<br clear="none">
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div><span>#else
/* !LJ_COMB_LB || CALC_ENERGIES */<br clear="none">
float
sig_r = sigma*inv_r;<br clear="none">
float
sig_r2 = sig_r*sig_r;<br clear="none">
float
sig_r6 = sig_r2*sig_r2*sig_r2;<br clear="none">
</span></div>
</div>
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div>And so on...</div>
<div><br clear="none">
</div>
<div>Please let me know if I still haven't
understand that.</div>
<br clear="none">
<div><br clear="none">
</div>
<div><br clear="none">
</div>
<div class="yiv0654810570gmail-m_-7076311669053638246ydp470688b9signature">Regards,<br clear="none">
Mahmood</div>
</div>
<div><br clear="none">
</div>
</div>
<div><br clear="none">
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</div>
<br clear="none">
<fieldset class="yiv0654810570mimeAttachmentHeader"></fieldset>
<br clear="none">
</div></div></div><div class="yqt4931552956" id="yqt77085">-- <br clear="none">Gromacs Developers mailing list<br clear="none"><br clear="none">* Please search the archive at <a shape="rect" href="http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List " target="_blank">http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List </a>before posting!<br clear="none"><br clear="none">* Can't post? Read <a shape="rect" href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br clear="none"><br clear="none">* For (un)subscribe requests visit<br clear="none"><a shape="rect" href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers " target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers </a>or send a mail to <a shape="rect" ymailto="mailto:gmx-developers-request@gromacs.org." href="mailto:gmx-developers-request@gromacs.org.">gmx-developers-request@gromacs.org.</a></div></div>
</div>
</div></body></html>