diff --git a/math/Makefile b/math/Makefile
index aa6f0c8ef2c2..343776406216 100644
--- a/math/Makefile
+++ b/math/Makefile
@@ -338,6 +338,7 @@
     SUBDIR += lll_spect
     SUBDIR += lp_solve
     SUBDIR += lrng
+    SUBDIR += lrslib
     SUBDIR += ltl
     SUBDIR += ltl2ba
     SUBDIR += lybniz
diff --git a/math/lrslib/Makefile b/math/lrslib/Makefile
new file mode 100644
index 000000000000..001cd2c089b6
--- /dev/null
+++ b/math/lrslib/Makefile
@@ -0,0 +1,31 @@
+# $FreeBSD$
+
+PORTNAME=	lrslib
+DISTVERSIONPREFIX=	${PORTNAME}-
+DISTVERSION=	062
+DISTVERSIONSUFFIX=	+autotools-2017-03-03
+CATEGORIES=	math
+
+MAINTAINER=	yuri@FreeBSD.org
+COMMENT=	Reverse search algorithm for vertex enumeration/convex hull problems
+
+LICENSE=	GPLv2
+LICENSE_FILE=	${WRKSRC}/COPYING
+
+LIB_DEPENDS=	libboost_system.so:devel/boost-libs \
+		libgmp.so:math/gmp \
+		libmpi.so:net/mpich
+
+USES=		autoreconf fortran gmake libtool localbase:ldflags # fortran because it's built with mpic++
+GNU_CONFIGURE=	yes
+CONFIGURE_ARGS=	--disable-static
+USE_GITHUB=	yes
+GH_ACCOUNT=	mkoeppe
+USE_LDCONFIG=	yes
+
+LDFLAGS+=	-lthr
+
+post-stage:
+	@${RM} -r ${STAGEDIR}${DATADIR}
+
+.include <bsd.port.mk>
diff --git a/math/lrslib/distinfo b/math/lrslib/distinfo
new file mode 100644
index 000000000000..f729a9c7653b
--- /dev/null
+++ b/math/lrslib/distinfo
@@ -0,0 +1,3 @@
+TIMESTAMP = 1543994635
+SHA256 (mkoeppe-lrslib-lrslib-062+autotools-2017-03-03_GH0.tar.gz) = 23aeba557e87b3613fb93b277772c7239c68cfc08d9a6bc437a7b993b02ce5ad
+SIZE (mkoeppe-lrslib-lrslib-062+autotools-2017-03-03_GH0.tar.gz) = 301464
diff --git a/math/lrslib/pkg-descr b/math/lrslib/pkg-descr
new file mode 100644
index 000000000000..86da7424bff2
--- /dev/null
+++ b/math/lrslib/pkg-descr
@@ -0,0 +1,9 @@
+lrslib is a self-contained ANSI C implementation of the reverse search algorithm
+for  vertex enumeration/convex hull problems and comes with a choice of three
+arithmetic packages. Input file formats are compatible with Komei Fukuda's cdd
+package. All computations are done exactly in either multiple  precision or
+fixed integer arithmetic. Output is not stored in memory, so even problems with
+very large output sizes can sometimes be solved. The program is intended for
+Unix/Linux platforms, but will compile using gcc/cygwin on Windows.
+
+WWW: http://cgm.cs.mcgill.ca/~avis/C/lrs.html
diff --git a/math/lrslib/pkg-plist b/math/lrslib/pkg-plist
new file mode 100644
index 000000000000..00a135733d3e
--- /dev/null
+++ b/math/lrslib/pkg-plist
@@ -0,0 +1,18 @@
+bin/2nash
+bin/lrs
+bin/lrs1
+bin/lrsnash
+bin/mplrs
+bin/mplrs1
+bin/plrs
+bin/plrs1
+bin/plrsmp
+bin/redund
+bin/redund1
+bin/setnash
+bin/setnash2
+include/lrsgmp.h
+include/lrslib.h
+lib/liblrsgmp.so
+lib/liblrsgmp.so.0
+lib/liblrsgmp.so.0.0.0