CPFloat: Custom-Precision Floating-Point numbers
CPFloat is a C library for simulating low-precision floating-point arithmetics. CPFloat provides efficient routines for rounding, performing arithmetic operations, evaluating mathematical functions, and querying properties of the simulated low-precision format. Internally, numbers are stored in float
or double
arrays. The low-precision format (target format) follows an extension of the formats defined in the IEEE 754 standard [5] and is entirely specified by four parameters:
a positive integer p, which represents the number of digits of precision;
a positive integer emin, which represents the minimum supported exponent;
a positive integer emax, which represents the maximum supported exponent; and
a Boolean variable σ, set to true if subnormal are supported and to false otherwise.
Valid choices of p, emin, and emax depend on the format in which the converted numbers are to be stored (storage format). A more extensive description of the characteristics of the low-precision formats that can be used, together with more details on admissible values for p, emin, emax, and σ can be found in [1].
The library was originally intended as a faster version of the MATLAB function chop
[2], which is available on GitHub.
The latest versions of the library have a variety of subtle differences compared with chop
.
Since 14 June 2022,
chop
supports specifying the function for generating random numbers. The MEX interface of CPFloat does not offer this capability, as the pseudo-random numbers used are generated in C and not in MATLAB.Since v0.6.0, CPFloat allows users to specify emin and emax separately. In earlier versions, users can only specify emax, while emin is set to 1 – emax.
Since v0.6.0, the default 8-bit format
E4M3
has emax = 8 and emin = –6, which is consistent with the homonymous format in the December 2023 revision of the OCP 8-bit Floating Point Specification (OFP8) [3]. Inchop
, emax = 7 and emin = –6.
The code to reproduce the results of the tests in [1] is available on GitHub.
Dependencies
The only (optional) dependency of CPFloat is the C implementation of the PCG Library, which provides a variety of high-quality pseudo-random number generators. For an in-depth discussion of the algorithms underlying the PCG Library, we recommend the paper by Melissa O’Neill [4]. If the header file pcg_variants.h
in include/pcg-c/include/pcg_variants.h
is not included at compile-time with the --include
option, then CPFloat relies on the default C pseudo-random number generator.
The PCG Library is free software (see the Licensing information below), and its generators are more efficient, reliable, and flexible than any combination of the functions srand
, rand
, and rand_r
from the C standard library. A warning is issued at compile time if the location of pcg_variant.h
is not specified correctly.
Compiling the MEX interface requires a reasonably recent version of MATLAB or Octave.
Developer dependencies
Testing the MEX interface requires the function float_params
, which is available on GitHub. The unit tests for the C implementation in test/cpfloat_test.ts
require the check unit testing framework for C, including the ``checkmk` <https://github.com/libcheck/check/tree/master/checkmk>`_ script, and the subunit protocol.
Installation
No installation is needed in order to use CPFloat as a header-only library. The shared and static libraries can be built with
make lib
If the compilation is successful, the header and library files of CPFloat will be located in the build/include
and build/lib
directories, respectively.
The library can be installed in <path>
with
make install --prefix=<path>
which copies the header and library files in <path>/include
and <path>/lib
, respectively.
The default value of <path>
, which is used if the --prefix
option is not supplied, is /usr/local
.
MEX interface
The MEX interface can be compiled automatically with either
make mexmat # Compile MEX interface for MATLAB.
or
make mexoct # Compile MEX interface for Octave.
These two commands compile and autotune the MEX interface in MATLAB and Octave, respectively, by using the functions mex/cpfloat_compile.m
and mex/cpfloat_autotune.m
. To use the interface, the bin/
folder must be in MATLAB’s search path.
On a system where the make
build automation tool is not available, we recommend building the MEX interface by running the script cpfloat_compile_nomake.m
in the mex/
folder. The script attempts to compile and autotune the MEX interface using the default C compiler. The following code will download the repository as a ZIP file, inflate it, and try to compile it:
zip_url = 'https://codeload.github.com/north-numerical-computing/cpfloat/zip/refs/heads/main';
unzip(zip_url);
movefile('cpfloat-main', 'cpfloat')
cd('cpfloat/mex');
cpfloat_compile_nomake
A different compiler can be used by setting the value of the variable compilerpath
appropriately.
If the chosen compiler does not support OpenMP, only the sequential version of the algorithm will be produced and no autotuning will take place.
On Windows, we have not been able to compile the PCG Library using the C compiler recommended by MATLAB. Therefore, the script uses the pseudo-random number generator in the C standard library by default.
Autotuning
CPFloat provides a sequential and a parallel implementation of the rounding functions. OpenMP introduces some overhead, and using a single thread is typically faster for arrays with few elements. Therefore, the library provides a facility to switch between the single-threaded and the multi-threaded variants automatically, depending on the size of the input. The threshold is machine-dependent, and the best value for a given system can be found by invoking
make autotune
which compiles the file src/cpfloat_autotune.c
, runs it, and updates the files src/cpfloat_threshold_binary32.h
and src/cpfloat_threshold_binary64.h
. This procedure is run automatically when building the shared and static libraries.
Documentation
The documentation of CPFloat can be generated with the command
make docs
which relies on Doxygen to format the Javadoc-style comments in the source files, and on Sphinx, with the Breathe and Exhale extensions, to generate the HTML version of the documentation that can be found in the docs/html/
directory.
Using CPFloat
CPFloat can be used as a header-only, shared, or static library. Examples for these three scenarios can be found in the Makefile
(cf. targets $(BINDIR)cpfloat_test
, $(BINDIR)libcpfloat_shared_test
, and $(BINDIR)libcpfloat_static_test
, respectively). Here we provide a brief summary.
Header-only library. The only requirement is that the files in the
src/
directory be in the include path of the compiler. In order to use the PCG Library, one can either:specify the path of the file
pcg_variants.h
using the preprocessor option--include
(see the variableCFLAGS
in theMakefile
for an example); ormake sure that
pcg_variants.h
is in the include path and uncomment the preprocessor instruction on line 34 ofsrc/cpfloat_definitions.h
, that is,/* #include "pcg_variants.h" */
. In either case, it is necessary link the executable against thepcg-random
library, which can be obtained by passing the option-lpcg-random
to the linker. The librarylibpcg-random.a
must be in the load path.
Shared library. The five header files in the
build/include
directory must be in the include path of the compiler. The options-lcpfloat
and-lm
must be passed to the linker, and the librarieslibcpfloat.so
andm.so
must be in the load path.Static library. The static library uses the same five header files as the shared library, which are located in the
build/include
and must be in the include path of the compiler. Executable must be linked with the-static
and-lcpfloat
options, and the library filelibcpfloat.a
must be in the load path. Linking against the math library is not needed in this case.
Code validation
The test/
directory contains two sets of test, one for the C library and one for the MEX interface. The unit tests for the C implementation require the check
library, and can be run with
make ctest
for the header-only library or with
make libtest
for the shared and static libraries. The two commands use the same batch of unit tests, which is generated from the file test/cpfloat_test.ts
using the checkmk
script.
The Makefile target coverage
measures the code coverage using GNU gcov
on the same set of tests.
The MEX interface can be tested by using either
make mtest # Test MEX interface using MATLAB.
or
make otest # Test MEX interface using Octave.
These two commands run, in MATLAB and Octave respectively, the function test/cpfloat_test.m
. This set of tests is based on the MATLAB script test_chop.m
, available on GitHub: some changes were necessary in order to make it compatible with Octave.
References
[1] Massimiliano Fasi and Mantas Mikaitis. CPFloat: A C library for simulating low-precision arithmetic. ACM Trans. Math. Softw., 49(2), Article No.: 18, June 2023.
[2] Nicholas J. Higham and Srikara Pranesh, Simulating Low Precision Floating-Point Arithmetic, SIAM J. Sci. Comput., 41, C585-C602, 2019.
[3] Paulius Micikevicius, Stuart Oberman, Pradeep Dubey, Marius Cornea, Andres Rodriguez, Ian Bratt, Richard Grisenthwaite, Norm Jouppi, Chiachen Chou, Amber Huffman, Michael Schulte, Ralph Wittig, Dharmesh Jani, Summer Deng. OCP 8-bit Floating Point Specification (OFP8), pp. 1–16, Revision 1.0, Open Compute Project, June 2023. Revised December 2023.
[4] Melissa E. O’Neill, PCG: A family of simple fast space-efficient statistically good algorithms for random number generation, Technical report HMC-CS-2014-0905, Harvey Mudd College, Claremont, CA, September 2014.
[5] 754-2019 IEEE Standard for Floating-Point Arithmetic, pp. 1–84, Institute of Electrical and Electronics Engineers, July 2019. Revision of IEEE Std 754-2008.
Acknowledgements
The library was written by Massimiliano Fasi and Mantas Mikaitis. We thank Theo Mary, Ian McInerney, and Siegfried Rump for reporting bugs and suggesting improvements.
Licensing information
CPFloat is distributed under the GNU Lesser General Public License, Version 2.1 or later (see LICENSE.md). Please contact us if you would like to use CPFloat in an open source project distributed under the terms of a license that is incompatible with the GNU LGPL. We might be able to relicense the software for you.
The PCG Library is distributed under the terms of either the Apache License, Version 2.0 or the Expat License, at the option of the user.
The MATLAB function float_params
is distributed under the terms of the BSD 2-Clause “Simplified” License.
The MATLAB function chop
is distributed under the terms of the BSD 2-Clause “Simplified” License.