<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
</head>
<body>
<div dir="auto" class="PlainText" style="font-size: 11pt;">You are correct, I meant to type 2025.3.0, but it turns out that it is ifx version 2025.3.2.<br>
<br>
I do not claim to be an expert on compilers, but my understanding is that the -march flags should still work because the new ifx/icx/icpx compilers are based on clang, which contains architecture specific optimizations for both Intel and AMD processors. Zen5
can take full 512bit AVX512 vectors when Zen4 cannot always do so, for instance.</div>
<div dir="auto" class="PlainText" style="font-size: 11pt;"><br>
</div>
<div dir="auto" class="PlainText" style="font-size: 11pt;">Benchmarking these flags would take more free time than I have.</div>
<div dir="auto" class="PlainText" style="font-size: 11pt;"><br>
</div>
<div dir="auto" class="PlainText" style="font-size: 11pt;">That said, on an identical WIEN2k job, a 16 core AMD 9950x is a little faster than 80 cores of Xeon E5-2680v2 spread out across 4 nodes (two 10 core cpus per node). <br>
<br>
<br>
Daniel Straus<br>
Assistant Professor<br>
Department of Chemistry<br>
Tulane University<br>
5088 Percival Stern Hall<br>
6400 Freret Street<br>
New Orleans, LA 70118<br>
(504) 862-3585<br>
<a href="http://straus.tulane.edu/">http://straus.tulane.edu/</a><br>
<br>
<br>
-----Original Message-----<br>
From: Wien <wien-bounces@zeus.theochem.tuwien.ac.at> On Behalf Of Fecher, Gerhard<br>
Sent: Friday, March 13, 2026 3:30 AM<br>
To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at><br>
Subject: Re: [Wien] Issues with -O3 optimization and ifx<br>
<br>
External Sender. Be aware of links, attachments and requests.<br>
<br>
Hallo Daniel,<br>
is the ifx version correct ? was fftw3 (and other libraries) compiled with the same compiler version ?<br>
ifx -v should tell: ifx version 2025.3.2<br>
<br>
I compiled Wien2k and fftw 3 succesfully with that version, my Options are:<br>
free -w -O3 -axCORE-AVX512 -fp-model=precise -DINTEL_VML -traceback -assume buffered_io -I$(MKLROOT)/include<br>
<br>
I had no problems to run Wien2k on a machine with 2x AMD EPYC 9354, that is the results of a fast test were the same as with ifort.<br>
<br>
however, -O3 was never recommended and it does not accelerate anything, usually I use -O2 the switch -axCORE-AVX512 is also not needed (even so sometimes mentioned for AMD processors), ... on Intel XEONs I use -xHost which brings some performance if OMP extensions
are used.<br>
I would recommend for the AMDs:<br>
free -w -O2 -fp-model=precise -DINTEL_VML -traceback -assume buffered_io -I$(MKLROOT)/include<br>
<br>
It seems that the switch march=xnver4 or xnver4 is only used for compatibility with the gnu compiler.<br>
I mentioned already earlier that there is no need for Intel to optimize the programs for AMD processors.<br>
<br>
I made some tests with openmp and SIMD extensions using a programm with parallelized loops, there was no improvement of performance with this switch, the fastes runs have been always those where I used only -O2, in some OMP cases znver4 resulted in a performance
loss for double precision calculations.<br>
<br>
The znver4 switch did also not improve when used with the AMD compiler (flang), however the flang with -O2 resulted (compared to ifx) in faster execution of the SIMD test.<br>
<br>
Ciao<br>
Gerhard<br>
<br>
DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:<br>
"I think the problem, to be quite honest with you, is that you have never actually known what the question is."<br>
<br>
====================================<br>
Dr. Gerhard H. Fecher<br>
Institut of Physics<br>
Johannes Gutenberg - University<br>
55099 Mainz<br>
________________________________________<br>
Von: Wien [wien-bounces@zeus.theochem.tuwien.ac.at] im Auftrag von Straus, Daniel B [dstraus@tulane.edu]<br>
Gesendet: Donnerstag, 12. März 2026 21:17<br>
An: A Mailing list for WIEN2k users<br>
Betreff: [Wien] Issues with -O3 optimization and ifx<br>
<br>
On a recently compiled copy of version 24.1 (with all patches applied) using Intel ifx version 2025.30, I attempted to set up a calculation but encountered very strange issues with nn where it gave nonsensical distances between atoms. On an older computer where
ifort was used to compile WIEN2k 24.1, the same cif worked fine in cif2struct, setrmt, nn, etc.<br>
<br>
I tracked it down to a compiler flag issue related to using the -O3 optimization flag. Using -O instead of -O3 solved the problem, and now everything behaves as expected.<br>
<br>
For what it is worth, my successful compiler flags are (the system has a Ryzen 9950x, hence the march=znver5 flag):<br>
(for regular execution) -O -march=znver5 -traceback -assume buffered_io -FR -I$(MKLROOT)/include -ipo -fp-model=precise (for parallel execution, linked to ELPA) -O -march=znver5 -traceback -assume buffered_io -FR -I$(MKLROOT)/include -ipo -fp-model=precise
-static -fc=ifx<br>
<br>
Hopefully, this message can save someone a few hours of troubleshooting down the line.<br>
<br>
<br>
Daniel Straus<br>
Assistant Professor<br>
Department of Chemistry<br>
Tulane University<br>
5088 Percival Stern Hall<br>
6400 Freret Street<br>
New Orleans, LA 70118<br>
(504) 862-3585<br>
<a href="http://straus.tulane.edu/">http://straus.tulane.edu/</a><br>
<br>
<br>
_______________________________________________<br>
Wien mailing list<br>
Wien@zeus.theochem.tuwien.ac.at<br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien">https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fzeus.theochem.tuwien.ac.at%2Fmailman%2Flistinfo%2Fwien&data=05%7C02%7Cdstraus%40tulane.edu%7C7cfe6c1a966c4cbeca0208de80daba89%7C9de9818325d94b139fc34de5489c1f3b%7C0%7C0%7C639089874038603736%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=byK3Iw232OJ6NOQeAg5LQmrCWCV3v0rr%2Brsc3T4K5rE%3D&reserved=0</a><br>
SEARCH the MAILING-LIST at: <a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html">
https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.mail-archive.com%2Fwien%40zeus.theochem.tuwien.ac.at%2Findex.html&data=05%7C02%7Cdstraus%40tulane.edu%7C7cfe6c1a966c4cbeca0208de80daba89%7C9de9818325d94b139fc34de5489c1f3b%7C0%7C0%7C639089874038651591%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=drmbhOiWF5EpYVIfsjzJA1x1nUONWz8enGvOZPvFsGo%3D&reserved=0</a><br>
</div>
</body>
</html>