blif support for unsigned magnitude comparison.

tgt-blif/Makefile.in

@@ -44,7 +44,8 @@ CPPFLAGS = $(INCLUDE_PATH) @CPPFLAGS@ @DEFS@ @PICFLAG@
 CXXFLAGS = @WARNING_FLAGS@ @CXXFLAGS@
 LDFLAGS = @LDFLAGS@
 
-O = blif.o constants.o logic_gate.o lpm.o lpm_add.o lpm_cmp_eq.o nex_data.o
+O = blif.o constants.o logic_gate.o lpm.o lpm_add.o lpm_cmp_eq.o lpm_cmp_gt.o \
+    nex_data.o
 
 all: dep blif.tgt
 

tgt-blif/lpm.cc

@@ -107,6 +107,10 @@ int print_lpm(FILE*fd, ivl_lpm_t net)
 	  case IVL_LPM_CMP_EEQ:
 	    rc += print_lpm_cmp_eq(fd, net);
 	    break;
+	  case IVL_LPM_CMP_GE:
+	  case IVL_LPM_CMP_GT:
+	    rc += print_lpm_cmp_gt(fd, net);
+	    break;
 	  case IVL_LPM_CONCAT:
 	  case IVL_LPM_CONCATZ:
 	    rc += print_concat(fd, net);

tgt-blif/lpm_cmp_gt.cc

@@ -0,0 +1,154 @@
+/*
+ * Copyright (c) 2013 Stephen Williams (steve@icarus.com)
+ *
+ *    This source code is free software; you can redistribute it
+ *    and/or modify it in source code form under the terms of the GNU
+ *    General Public License as published by the Free Software
+ *    Foundation; either version 2 of the License, or (at your option)
+ *    any later version.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    GNU General Public License for more details.
+ *
+ *    You should have received a copy of the GNU General Public License
+ *    along with this program; if not, write to the Free Software
+ *    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ */
+
+# include  "priv.h"
+# include  "nex_data.h"
+# include  <cassert>
+
+static int print_lpm_cmp_gt_s(FILE*fd, ivl_lpm_t net)
+{
+      ivl_nexus_t q_nex = ivl_lpm_q(net);
+      ivl_nexus_t a_nex = ivl_lpm_data(net,0);
+      ivl_nexus_t b_nex = ivl_lpm_data(net,1);
+
+      blif_nex_data_t*q_ned = blif_nex_data_t::get_nex_data(q_nex);
+      blif_nex_data_t*a_ned = blif_nex_data_t::get_nex_data(a_nex);
+      blif_nex_data_t*b_ned = blif_nex_data_t::get_nex_data(b_nex);
+
+      assert(1                  == q_ned->get_width());
+      assert(ivl_lpm_width(net) == a_ned->get_width());
+      assert(ivl_lpm_width(net) == b_ned->get_width());
+
+	// This is true if the operator is GE instead of just GT.
+      bool ge_flag = ivl_lpm_type(net)==IVL_LPM_CMP_GE;
+
+	// Special case: if the input vector is a single bit, then the
+	// operation is trivial. Note that since this is signed, the
+	// "1" bit represents -1 and is <0.
+      if (ivl_lpm_width(net) == 1) {
+	    fprintf(fd, ".names %s%s %s%s %s%s\n"
+		    "01 1\n",
+		    a_ned->get_name(), a_ned->get_name_index(0),
+		    b_ned->get_name(), b_ned->get_name_index(0),
+		    q_ned->get_name(), q_ned->get_name_index(0));
+	    if (ge_flag) fprintf(fd, "00 1\n11 1\n");
+	    return 0;
+      }
+
+      fprintf(stderr, "%s:%u: sorry: blif: Signed agnitude compare not implemented yet\n",
+	      ivl_lpm_file(net), ivl_lpm_lineno(net));
+
+      return 1;
+}
+
+static int print_lpm_cmp_gt_u(FILE*fd, ivl_lpm_t net)
+{
+      ivl_nexus_t q_nex = ivl_lpm_q(net);
+      ivl_nexus_t a_nex = ivl_lpm_data(net,0);
+      ivl_nexus_t b_nex = ivl_lpm_data(net,1);
+
+      blif_nex_data_t*q_ned = blif_nex_data_t::get_nex_data(q_nex);
+      blif_nex_data_t*a_ned = blif_nex_data_t::get_nex_data(a_nex);
+      blif_nex_data_t*b_ned = blif_nex_data_t::get_nex_data(b_nex);
+
+      assert(1                  == q_ned->get_width());
+      assert(ivl_lpm_width(net) == a_ned->get_width());
+      assert(ivl_lpm_width(net) == b_ned->get_width());
+
+	// This is true if the operator is GE instead of just GT.
+      bool ge_flag = ivl_lpm_type(net)==IVL_LPM_CMP_GE;
+
+	// Special case: if the input vector is a single bit, then the
+	// operation is trivial.
+      if (ivl_lpm_width(net) == 1) {
+	    fprintf(fd, ".names %s%s %s%s %s%s\n"
+		    "10 1\n",
+		    a_ned->get_name(), a_ned->get_name_index(0),
+		    b_ned->get_name(), b_ned->get_name_index(0),
+		    q_ned->get_name(), q_ned->get_name_index(0));
+	    if (ge_flag) fprintf(fd, "00 1\n11 1\n");
+	    return 0;
+      }
+
+	// Calculate GT and EQ for each bit.
+      for (unsigned idx = 0 ; idx < ivl_lpm_width(net) ; idx += 1) {
+	    fprintf(fd, ".names %s%s %s%s %s%s/G%u\n"
+		    "10 1\n",
+		    a_ned->get_name(), a_ned->get_name_index(idx),
+		    b_ned->get_name(), b_ned->get_name_index(idx),
+		    q_ned->get_name(), q_ned->get_name_index(0), idx);
+      }
+      for (unsigned idx = ge_flag?0 : 1 ; idx < ivl_lpm_width(net) ; idx += 1) {
+	    fprintf(fd, ".names %s%s %s%s %s%s/E%u\n"
+		    "11 1\n"
+		    "00 1\n",
+		    a_ned->get_name(), a_ned->get_name_index(idx),
+		    b_ned->get_name(), b_ned->get_name_index(idx),
+		    q_ned->get_name(), q_ned->get_name_index(0), idx);
+      }
+
+	// Generate a wide function to blend the per-bit comparisons.
+      fprintf(fd, ".names");
+      for (unsigned idx = 0 ; idx < ivl_lpm_width(net) ; idx += 1) {
+	    fprintf(fd, " %s%s/G%u",
+		    q_ned->get_name(), q_ned->get_name_index(0), idx);
+      }
+      for (unsigned idx = ge_flag?0:1 ; idx < ivl_lpm_width(net) ; idx += 1) {
+	    fprintf(fd, " %s%s/E%u",
+		    q_ned->get_name(), q_ned->get_name_index(0), idx);
+      }
+      fprintf(fd, " %s%s\n", q_ned->get_name(), q_ned->get_name_index(0));
+
+      for (unsigned idx = 0 ; idx < ivl_lpm_width(net) ; idx += 1) {
+	    for (unsigned jdx = 0 ; jdx < ivl_lpm_width(net) ; jdx += 1) {
+		  if (jdx != idx)
+			fprintf(fd, "-");
+		  else
+			fprintf(fd, "1");
+	    }
+	    for (unsigned jdx = ge_flag?0:1 ; jdx < ivl_lpm_width(net) ; jdx += 1) {
+		  if (jdx <= idx)
+			fprintf(fd, "-");
+		  else
+			fprintf(fd, "1");
+	    }
+	    fprintf(fd, " 1\n");
+      }
+
+      if (ge_flag) {
+	    for (unsigned idx = 0 ; idx < ivl_lpm_width(net) ; idx += 1)
+		  fputc('-', fd);
+	    for (unsigned idx = 0 ; idx < ivl_lpm_width(net) ; idx += 1)
+		  fputc('1', fd);
+	    fprintf(fd, " 1\n");
+      }
+
+      return 0;
+}
+
+int print_lpm_cmp_gt(FILE*fd, ivl_lpm_t net)
+{
+      fprintf(fd, "# %s:%u: LPM_LPM_CMP_GT: width=%u\n",
+	      ivl_lpm_file(net), ivl_lpm_lineno(net), ivl_lpm_width(net));
+
+      if (ivl_lpm_signed(net))
+	    return print_lpm_cmp_gt_s(fd, net);
+      else
+	    return print_lpm_cmp_gt_u(fd, net);
+}

tgt-blif/priv.h

@@ -34,6 +34,7 @@ extern int print_lpm(FILE*fd, ivl_lpm_t net);
 extern int print_lpm_add(FILE*fd, ivl_lpm_t net);
 extern int print_lpm_sub(FILE*fd, ivl_lpm_t net);
 extern int print_lpm_cmp_eq(FILE*fd, ivl_lpm_t net);
+extern int print_lpm_cmp_gt(FILE*fd, ivl_lpm_t net);
 extern int print_lpm_cmp_ne(FILE*fd, ivl_lpm_t net);
 
 /*