Fix: update_field_class_in_ir(): do not have special int fields in trace IR

The logic in update_field_class_in_ir() for the integer field class case
was wrong. This patch fixes this, so that we only translate the CTF IR
integer field class to a trace IR integer field class if another field
class depends on it (dynamic array, variant) or if it does not update a
clock AND it does not have a special meaning.

Signed-off-by: Philippe Proulx <eeppeliteloop@gmail.com>

diff --git a/plugins/ctf/common/metadata/ctf-meta-update-in-ir.c b/plugins/ctf/common/metadata/ctf-meta-update-in-ir.c
index ed5d23bfd25b060b0311dae7e81df95ac2ed0d0a..0cb95cf7a82a9b1b124cfbd66eab94e59e16a2db 100644 (file)
--- a/plugins/ctf/common/metadata/ctf-meta-update-in-ir.c
+++ b/plugins/ctf/common/metadata/ctf-meta-update-in-ir.c
@@ -99,14 +99,16 @@ void update_field_class_in_ir(struct ctf_field_class *fc,
        {
                struct ctf_field_class_int *int_fc = (void *) fc;
 
-               if (int_fc->mapped_clock_class ||
-                               int_fc->meaning == CTF_FIELD_CLASS_MEANING_NONE ||
+               /*
+                * Conditions to be in trace IR; one of:
+                *
+                * 1. Does NOT have a mapped clock class AND does not
+                *    have a special meaning.
+                * 2. Another field class depends on it.
+                */
+               if ((!int_fc->mapped_clock_class &&
+                               int_fc->meaning == CTF_FIELD_CLASS_MEANING_NONE) ||
                                bt_g_hash_table_contains(ft_dependents, fc)) {
-                       /*
-                        * Field class does not update a clock, has no
-                        * special meaning, and no sequence/variant
-                        * field class which is part of IR depends on it.
-                        */
                        fc->in_ir = true;
                }