Defining new debug utils for v4dualfloat64 precision

nibrunie · Nov 18, 2018 · 46042d0 · 46042d0
1 parent ee07120
commit 46042d0
Showing 1 changed file with 5 additions and 0 deletions.
diff --git a/metalibm_core/utility/debug_utils.py b/metalibm_core/utility/debug_utils.py
@@ -94,6 +94,9 @@ def fixed_point_pre_process(value, optree):
 debug_ulong4  = ML_Debug(display_format = "{%lu, %lu, %lu, %lu}", pre_process = lambda v: "%s._[0], %s._[1], %s._[2], %s._[3]" % (v, v, v, v))
 debug_ulong8  = ML_Debug(display_format = "{%lu, %lu, %lu, %lu, %lu, %lu, %lu, %lu}", pre_process = lambda v: "%s._[0], %s._[1], %s._[2], %s._[3], %s._[4], %s._[5], %s._[6], %s._[7]" % (v, v, v, v, v, v, v, v))
 
+
+debug_v4dualfloat64 = ML_Debug(display_format="[{%a, %a}, {%a, %a}, {%a, %a}, {%a, %a}]", pre_process=lambda v: (", ".join("{v}.hi._[%d], {v}.lo._[%d]" % (i, i) for i in range(4))).format(v=v)) 
+
 debug_multi = ML_MultiDebug({
   ML_Binary32: debug_ftox,
   ML_Binary64: debug_lftolx,
@@ -135,4 +138,6 @@ def fixed_point_pre_process(value, optree):
 
   ML_TripleDouble: debug_td,
   ML_TripleSingle: debug_ts,
+
+  v4dualfloat64: debug_v4dualfloat64,
 })