regcomp.c: Fix more alignment problems

I believe this will fix the remaining alignment problems recently being
shown on gcc on HP-UX, It works on the procura machine.

regnodes should not have stricter alignment than required by U32, for
reasons given in the comments this commit adds to the beginning of
regcomp.h.  Commit 31f05a37 added a new ANYOF regnode struct with a
pointer field.  This requires stricter alignment on some 64-bit platforms,
and hence doesn't work on those platforms.

This commit removes that regnode struct type, and instead stores the
pointer it used via a more indirect, but already existing mechanism
that stores other data..

The function that returns that other data is enlarged to return this new
field as well.  It now needs to be called from regcomp.c, so the
previous commit had renamed and made it accessible from there.  The
"public" function that wraps this one is unchanged.  (I put "public" in
quotes here, because I don't think anyone outside core is or should be
using it, but since it has been publicly available for a long time, I'm
treating the API as unchangeable.  regcomp.c called this public function
before this commit, but needs the additional data returned by the inner
one).

1 files changed, 16 insertions, 20 deletions

diff --git a/regcomp.h b/regcomp.h
index 744f361dc4..700d6c197d 100644
--- a/regcomp.h
+++ b/regcomp.h
@@ -187,6 +187,17 @@ struct regnode_2 {
 
 #define ANYOF_BITMAP_SIZE	(256 / 8)   /* 8 bits/Byte */
 
+/* Note that these form structs which are supersets of the next smaller one, by
+ * appending fields.  Alignment problems can occur if one of those optional
+ * fields requires stricter alignment than the base struct.  And formal
+ * parameters that can really be two or more of the structs should be
+ * declared as the smallest one it could be.  See commit message for
+ * 7dcac5f6a5195002b55c935ee1d67f67e1df280b.  Regnode allocation is done
+ * without regard to alignment, and changing it to would also require changing
+ * the code that inserts and deletes regnodes.  The basic single-argument
+ * regnode has a U32, which is what reganode() allocates as a unit.  Therefore
+ * no field can require stricter alignment than U32. */
+
 /* also used by trie */
 struct regnode_charclass {
     U8	flags;
@@ -206,23 +217,14 @@ struct regnode_charclass_class {
     U32 classflags;	                        /* and run-time */
 };
 
-/* like above, but also has folds that are used only if the runtime locale is
- * UTF-8. */
-struct regnode_charclass_posixl_fold {
-    U8	flags;				/* ANYOF_POSIXL bit must go here */
-    U8  type;
-    U16 next_off;
-    U32 arg1;
-    char bitmap[ANYOF_BITMAP_SIZE];	/* both compile-time */
-    U32 classflags;	                /* and run-time */
-    SV* utf8_locale_list;               /* list of code points matched by folds
-                                           in a UTF-8 locale */
-};
-
 /* A synthetic start class; is a regnode_charclass_posixl_fold, plus an extra
  * SV*, used only during its construction and which is not used by regexec.c.
  * Note that the 'next_off' field is unused, as the SSC stands alone, so there
- * is never a next node. */
+ * is never a next node.  Also, there is no alignment issue, becase these are
+ * declared or allocated as a complete unit so the compiler takes care of
+ * alignment.  This is unlike ithe other regnodes which are allocated in terms
+ * of multiples of a single-argument regnode.  Because there is no alignment
+ * issue, these can have a pointer field */
 struct regnode_ssc {
     U8	flags;				/* ANYOF_POSIXL bit must go here */
     U8  type;
@@ -230,8 +232,6 @@ struct regnode_ssc {
     U32 arg1;
     char bitmap[ANYOF_BITMAP_SIZE];	/* both compile-time */
     U32 classflags;	                /* and run-time */
-    SV* utf8_locale_list;               /* list of code points matched by folds
-                                           in a UTF-8 locale */
     SV* invlist;                        /* list of code points matched */
 };
 
@@ -492,7 +492,6 @@ struct regnode_ssc {
 #define ANYOF_SIZE		(sizeof(struct regnode_charclass))
 #define ANYOF_POSIXL_SIZE	(sizeof(regnode_charclass_posixl))
 #define ANYOF_CLASS_SIZE	ANYOF_POSIXL_SIZE
-#define ANYOF_POSIXL_FOLD_SIZE  (sizeof(regnode_charclass_posixl_fold))
 
 #define ANYOF_FLAGS(p)		((p)->flags)
 
@@ -554,11 +553,8 @@ struct regnode_ssc {
 
 #define ANYOF_SKIP		((ANYOF_SIZE - 1)/sizeof(regnode))
 #define ANYOF_POSIXL_SKIP	((ANYOF_POSIXL_SIZE - 1)/sizeof(regnode))
-#define ANYOF_POSIXL_FOLD_SKIP  ((ANYOF_POSIXL_FOLD_SIZE - 1)/sizeof(regnode))
 #define ANYOF_CLASS_SKIP	ANYOF_POSIXL_SKIP
 
-#define ANYOF_UTF8_LOCALE_INVLIST(node) (((regnode_charclass_posixl_fold*) (node))->utf8_locale_list)
-
 /*
  * Utility definitions.
  */