<div dir="ltr"><br><br><div class="gmail_quote">On Wed, May 13, 2015 at 10:24 PM Berk Hess <<a href="mailto:hess@kth.se">hess@kth.se</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div>Hi,<br>
<br>
I also noticed only now that there is more logic on line 158 of
FindFFTW.cmake that probably should have been updated (as well as
the documentation at the top). I should have stayed out of the
business of editing cmake files :-(</div></div></blockquote><div><br></div><div>It's not exactly your fault that nobody thought to comment anywhere that FindFFTW.cmake and the own-fftw build are supposed to fulfil the same contract...</div><div><br></div><div>Anyway, bring on the day when we can do our own FFTs! ;-)</div><div><br></div><div>Mark</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div bgcolor="#FFFFFF" text="#000000"><div><br>
<br>
Berk</div></div><div bgcolor="#FFFFFF" text="#000000"><div><br>
<br>
On 05/13/2015 10:21 PM, Mark Abraham wrote:<br>
</div></div><div bgcolor="#FFFFFF" text="#000000">
<blockquote type="cite">
<div dir="ltr">Hi,<br>
<br>
<div class="gmail_quote">On Wed, May 13, 2015 at 8:54 PM Berk
Hess <<a href="mailto:hess@kth.se" target="_blank">hess@kth.se</a>>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div>Hi,<br>
<br>
I had a look at my change to the FFTW checks, but I
don't see how something can go wrong here. As far as I
can see the order and essential logic have not changed
compared to 5.0.4.<br>
</div>
</div>
</blockquote>
<div><br>
</div>
<div>The issue is on line 91. The new AND NOT works only under
the assumption that FFTWF_FOUND means that we ran the
SIMD-flavour trycompile tests. But the auto-download sets
FFTWF_FOUND *and* FFTWF_HAVE_SIMD (because it will have
SIMD) and runs/fakes none of the tests. The previous logic
was positive, so the issue did not arise. One fix is to
change src/contrib/fftw/CMakeLists.txt to set a fake
${FFTW}_HAVE_SSE2.</div>
<div><br>
</div>
<div>Ideally, we could have a sane test that the auto-download
works, but even if we'd had one, the build would have worked
and we likely wouldn't have verified the warning-issuing
behaviour :-(</div>
<div><br>
</div>
<div>Mark</div>
<div><br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div> Cheers,<br>
<br>
Berk</div>
</div>
<div bgcolor="#FFFFFF" text="#000000">
<div><br>
<br>
On 05/13/2015 05:54 PM, Mark Abraham wrote:<br>
</div>
</div>
<div bgcolor="#FFFFFF" text="#000000">
<blockquote type="cite">
<div dir="ltr">Hi,
<div><br>
</div>
<div>Thanks for the email.</div>
<div><br>
<div class="gmail_quote">On Wed, May 13, 2015 at
5:34 PM Hardy, Adam <<a href="mailto:ah259@hw.ac.uk" target="_blank">ah259@hw.ac.uk</a>>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear all,<br>
<br>
I'm having trouble installing the latest
release. Leaving all other options default, but
switching GMX_BUILD_OWN_FFTW=yes results in a
cmake warning:<br>
<br>
"<br>
CMake Warning at
cmake/gmxManageFFTLibraries.cmake:93 (message):<br>
The FFTW library was compiled with neither
--enable-sse nor --enable-sse2;<br>
those would have enabled SSE(2) SIMD
instructions. This will give<br>
suboptimal performance. You should
(re)compile the FFTW library with both<br>
SSE2 and AVX instruction support (use both
--enable-sse2 and --enable-avx).<br>
The FFTW library will determine at runtime
which SIMD instruction set is<br>
fastest for different parts of the FFTs.<br>
Call Stack (most recent call first):<br>
CMakeLists.txt:742 (include) </blockquote>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">"</blockquote>
<div> </div>
<div>That's inconsistent and somewhat embarassing,
but not actually a problem. FFTW hasn't been
downloaded or built by the time that warning is
issued, so we have some logic that doesn't work
correctly in the auto-download case.</div>
<div><br>
</div>
<div>Later on, FFTW is built but is still compiled
with only --enable-sse2. We may change that
behaviour for 5.1, if it is clear that adding
--enable-avx as well is always
neutral-or-useful, but there's nothing wrong
with the resulting mdrun here.</div>
<div><br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> This configuration
works as expected with 5.0.4.<br>
<br>
I get the same results with this setting as
"no". This message doesn't make much sense in
either case as I do not have an FFTW library on
my machine.</blockquote>
<div><br>
</div>
<div>Hmm that could be a minor problem, but none
of us have tried a build on a machine without
FFTW, for the obvious reason. :-)</div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">The error I get in 5.0.4
for this scenario is:<br>
<br>
"<br>
Could not find fftw3f library named libfftw3f,
please specify its location in CMAKE_PREFIX_PATH
or FFTWF_LIBRARY by hand (e.g.<br>
-DFFTWF_LIBRARY='/path/to/libfftw3f.so')<br>
<br>
CMake Error at
cmake/gmxManageFFTLibraries.cmake:76 (MESSAGE):<br>
Cannot find FFTW 3 (with correct precision -
libfftw3f for mixed-precision<br>
GROMACS or libfftw3 for double-precision
GROMACS). Either choose the right<br>
precision, choose another FFT(W) library
(-DGMX_FFT_LIBRARY), enable the<br>
advanced option to let GROMACS build FFTW 3
for you<br>
(-GMX_BUILD_OWN_FFTW=ON), or use the really
slow GROMACS built-in fftpack<br>
library (-DGMX_FFT_LIBRARY=fftpack).<br>
Call Stack (most recent call first):<br>
CMakeLists.txt:738 (include)<br>
"<br>
<br>
I see some in the release notes some changes
were made to the cmake warnings for AVX capable
processors which I belive mine is </blockquote>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">(AMD fx-8350).<br>
</blockquote>
<div><br>
</div>
<div>Yes, but we don't expect much/any improvement
on AMD cpus from adding --enable-avx. Use
--enable-sse2 always (which is what the
auto-download does).</div>
<div><br>
</div>
<div>Mark</div>
<div><br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> Thanks,<br>
<br>
Adam Hardy<br>
PhD Student<br>
School of Engineering and Physical Sciences<br>
Heriot-Watt University<br>
Edinburgh EH14 4AS, UK<br>
<br>
-----<br>
We invite research leaders and ambitious early
career researchers to<br>
join us in leading and driving research in key
inter-disciplinary themes.<br>
Please see <a href="http://www.hw.ac.uk/researchleaders" target="_blank">www.hw.ac.uk/researchleaders</a>
for further information and how<br>
to apply.<br>
<br>
Heriot-Watt University is a Scottish charity<br>
registered under charity number SC000278.<br>
<br>
--<br>
Gromacs Developers mailing list<br>
<br>
* Please search the archive at <a 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>
<br>
* Can't post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
<br>
* For (un)subscribe requests visit<br>
<a 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 href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a>.<br>
</blockquote>
</div>
</div>
</div>
<br>
<fieldset></fieldset>
<br>
</blockquote>
<br>
</div>
--<br>
Gromacs Developers mailing list<br>
<br>
* Please search the archive at <a 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>
<br>
* Can't post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
<br>
* For (un)subscribe requests visit<br>
<a 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 href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a>.</blockquote>
</div>
</div>
<br>
<fieldset></fieldset>
<br>
</blockquote>
<br>
</div>
--<br>
Gromacs Developers mailing list<br>
<br>
* Please search the archive at <a 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>
<br>
* Can't post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
<br>
* For (un)subscribe requests visit<br>
<a 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 href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a>.</blockquote></div></div>