Skip to content

Commit dad390c

Browse files
jwpetersonjwpeterson
committed
HTML docs updates.
These changes were originally made in libMesh/libMesh.github.io#2, but they need to made here as well so they aren't lost in the next website update.
1 parent 0d17f47 commit dad390c

File tree

2 files changed

+32
-51
lines changed

2 files changed

+32
-51
lines changed

‎doc/html/src/externalsoftware.html‎

Lines changed: 11 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,17 @@
11
<h1>External Packages</h1>
22
<code>libMesh</code> interfaces to a number of high-quality software packages to provide certain functionality. This page provides a list of packages and a description of their use in the library.
33

4-
<h2>MPI</h2> The <a href="http://www-unix.mcs.anl.gov/mpi">Message Passing Interface</a> is a standard for parallel programming using the message passing model. PETSc requires MPI for its functionality. <code>libMesh</code> makes use of MPI to when running in parallel for certain operations, including its <code>ParallelMesh</code> distributed-memory, fully unstructured mesh implementation.
4+
<h2>MPI</h2> The <a href="http://www-unix.mcs.anl.gov/mpi">Message
5+
Passing Interface</a> is a standard for parallel programming using
6+
the message passing model. PETSc and Trilinos require MPI for
7+
distributed-memory parallel functionality.
8+
<code>libMesh</code> can make use of MPI for certain operations on
9+
distributed-memory (as well as shared-memory and hybrid) parallel
10+
computers, enabling its <code>DistributedMesh</code>
11+
distributed-memory, fully unstructured mesh implementation.
12+
13+
<code>libMesh</code> currently only supports MPI implementations
14+
compatible with the MPI-2 and later MPI standards.
515

616
<h2>TBB</h2> Since February 2008 <code>libMesh</code> can be configured to use the <a href="http://threadingbuildingblocks.org/">Threading Building Blocks</a> for thread-based parallelism on shared memory machines. Several key algorithms in the library have been refactored to be multithreaded, and this effort will continue as additional profiling reveals additional serial bottlenecks. It is envisioned that for certain classes of problems multilevel parallelism (e.g. message passing between nodes and threading within nodes) will prove more scalable than message passing alone, especially with the introduction of commodity multi-core processors. The reality is that for implicit problems this can only be achieved with a parallel linear algebra library that also uses multilevel parallelism.
717

‎doc/html/src/installation.html‎

Lines changed: 21 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -27,60 +27,31 @@ <h1>Installation Instructions</h1>
2727

2828
<a id="compilers"></a><h2>Compilers</h2>
2929

30-
<code>libMesh</code> makes extensive use of the standard C++ library,
31-
so you will need a decent, standards-compliant compiler. We have tried
32-
very hard to make the code completely compiler-agnostic by avoiding
33-
questionable (but legal) constructs. If you have a compiler that won't
34-
build the code please let us know. You will also need a decent C compiler
30+
<code>libMesh</code> makes extensive use of the standard C++ library
31+
and increasing use of C++11 standard language and library features,
32+
so you will need a recent, standards-compliant compiler. We have tried
33+
very hard to make the code compiler-agnostic by avoiding questionable
34+
constructs. If you have a C++11-supporting compiler that won't
35+
build libMesh code please let us know. You will also need a decent C compiler
3536
if you want to build some of the contributed packages that add functionality
3637
to the library.
37-
38-
The library is continuously tested with the following compilers:
38+
<br>
3939

4040
<br>
41-
<ul>
42-
<li>GNU GCC
43-
<ul>
44-
<li><code>gcc</code> 3.4</li>
45-
<li><code>gcc</code> 4.2</li>
46-
<li><code>gcc</code> 4.4</li>
47-
<li><code>gcc</code> 4.5</li>
48-
<li><code>gcc</code> 4.6</li>
49-
<li><code>gcc</code> 4.7</li>
50-
<li><code>gcc</code> 4.8</li>
51-
</ul>
52-
</li>
53-
<li>Intel ICC
54-
<ul>
55-
<li><code>icc</code> 11.1</li>
56-
<li><code>icc</code> 12.0</li>
57-
<li><code>icc</code> 12.1</li>
58-
<li><code>icc</code> 13.0</li>
59-
</ul>
60-
</li>
61-
<li>Clang
62-
<ul>
63-
<li><code>clang++</code> 2.9</li>
64-
<li><code>clang++</code> 3.0</li>
65-
<li><code>clang++</code> 3.1</li>
66-
<li><code>clang++</code> 3.2</li>
67-
</ul>
68-
</li>
69-
<li>Sun Studio/Oracle
70-
<ul>
71-
<li><code>CC</code> 12.3</li>
72-
</ul>
73-
<code>$ ./configure --disable-fparser CXXFLAGS=-library=stlport4 --disable-unordered-containers</code>
74-
</li>
75-
<li> Portland Group
76-
<ul>
77-
<li><code>pgi</code> 11.7</li>
78-
<li><code>pgi</code> 12.9</li>
79-
<li><code>pgi</code> 13.4</li>
80-
</ul>
81-
<code>$ ./configure --disable-unordered-containers --disable-fparser --enable-static --disable-shared</code>
82-
</li>
83-
</ul>
41+
<code>libMesh</code> Continuous Integration testing of all library
42+
modifications currently is done with gcc 6.2 and gcc 7.3. Less
43+
frequent testing is done with clang, Intel, and older gcc compilers.
44+
45+
<a id="external"></a><h2>External Software</h2>
46+
47+
<code>libMesh</code> has many features which are enabled via
48+
integration with <a href="externalsoftware.html">various third-party
49+
libraries</a>, a few of which are redistributed in our contrib
50+
directory, others of which may be separately installed on your system.
51+
The libMesh configure script attempts to autodetect these libraries
52+
when possible. Some of the supported libraries have occasionally
53+
changing (or even frequently changing) APIs. We attempt to provide a
54+
range of backwards compatibility for old versions of third-party APIs.
8455

8556

8657
<a id="conf"></a><h2>Configuration</h2>

0 commit comments

Comments
 (0)