* function.c (purge_addressof_1): In (mem (addressof (reg))) case

	for reg notes, if there are no substitutions, just use a SUBREG.

git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@65934 138bc75d-0d04-0410-961f-82ee72b054a4

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index d6ffa22..9965b90 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,8 @@
+2003-04-22  J"orn Rennecke <joern.rennecke@superh.com>
+
+       * function.c (purge_addressof_1): In (mem (addressof (reg))) case
+       for reg notes, if there are no substitutions, just use a SUBREG.
+
 2003-04-21  Mark Mitchell  <mark@codesourcery.com>
 
        * Makefile.in (calls.o): Depend on except.h.

diff --git a/gcc/function.c b/gcc/function.c
index b75a07e..de10f73 100644 (file)
--- a/gcc/function.c
+++ b/gcc/function.c
@@ -3147,6 +3147,36 @@ purge_addressof_1 (loc, insn, force, store, may_postpone, ht)
                    return true;
                  }
 
+             /* When we are processing the REG_NOTES of the last instruction
+                of a libcall, there will be typically no replacements
+                for that insn; the replacements happened before, piecemeal
+                fashion.  OTOH we are not interested in the details of
+                this for the REG_EQUAL note, we want to know the big picture,
+                which can be succinctly described with a simple SUBREG.
+                Note that removing the REG_EQUAL note is not an option
+                on the last insn of a libcall, so we must do a replacement.  */
+             if (! purge_addressof_replacements
+                 && ! purge_bitfield_addressof_replacements)
+               {
+                 /* In compile/990107-1.c:7 compiled at -O1 -m1 for sh-elf,
+                    we got
+                    (mem:DI (addressof:SI (reg/v:DF 160) 159 0x401c8510)
+                     [0 S8 A32]), which can be expressed with a simple
+                    same-size subreg  */
+                 if ((GET_MODE_SIZE (GET_MODE (x))
+                      == GET_MODE_SIZE (GET_MODE (sub)))
+                     /* Again, invalid pointer casts (as in
+                        compile/990203-1.c) can require paradoxical
+                        subregs.  */
+                     || (GET_MODE_SIZE (GET_MODE (x)) > UNITS_PER_WORD
+                         && (GET_MODE_SIZE (GET_MODE (x))
+                             > GET_MODE_SIZE (GET_MODE (sub)))))
+                   {
+                     *loc = gen_rtx_SUBREG (GET_MODE (x), sub, 0);
+                     return true;
+                   }
+                 /* ??? Are there other cases we should handle?  */
+               }
              /* Sometimes we may not be able to find the replacement.  For
                 example when the original insn was a MEM in a wider mode,
                 and the note is part of a sign extension of a narrowed