re PR target/30778 (invalid code generation for memset() with -mtune=k8)

	PR target/30778
	* i386.c (counter_mode): New function.
	(expand_set_or_movmem_via_loop): Use it.
	(expand_movmem_epilogue): Likewise; fix pasto.
	(ix86_expand_movmem): Do emit guard even for constant counts.
	(ix86_expand_setmem): Likewise.

From-SVN: r122313

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 031c067eba921a45081722bcf898d389b8d86701..6aa21b5c158df614bfe3779bf87ff74a09d0db17 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,13 @@
+2007-02-24  Uros Bizjak  <ubizjak@gmail.com>
+           Jan Hubicka  <jh@suse.cz>
+
+       PR target/30778
+       * i386.c (counter_mode): New function.
+       (expand_set_or_movmem_via_loop): Use it.
+       (expand_movmem_epilogue): Likewise; fix pasto.
+       (ix86_expand_movmem): Do emit guard even for constant counts.
+       (ix86_expand_setmem): Likewise.
+
 2007-02-25  Nick Clifton  <nickc@redhat.com>
 
        * config/frv/frv.h (ASM_OUTPUT_CASE_LABEL): Delete.

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index db074413086e1602caeafb7da66bf9014cc8d9e5..5a4e1d5e4c6c68449e46d75f63058a6511b9177e 100644 (file)
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -13314,6 +13314,21 @@ scale_counter (rtx countreg, int scale)
   return sc;
 }
 
+/* Return mode for the memcpy/memset loop counter.  Preffer SImode over DImode
+   for constant loop counts.  */
+
+static enum machine_mode
+counter_mode (rtx count_exp)
+{
+  if (GET_MODE (count_exp) != VOIDmode)
+    return GET_MODE (count_exp);
+  if (GET_CODE (count_exp) != CONST_INT)
+    return Pmode;
+  if (TARGET_64BIT && (INTVAL (count_exp) & ~0xffffffff))
+    return DImode;
+  return SImode;
+}
+
 /* When SRCPTR is non-NULL, output simple loop to move memory
    pointer to SRCPTR to DESTPTR via chunks of MODE unrolled UNROLL times,
    overall size is COUNT specified in bytes.  When SRCPTR is NULL, output the
@@ -13330,7 +13345,7 @@ expand_set_or_movmem_via_loop (rtx destmem, rtx srcmem,
                               int expected_size)
 {
   rtx out_label, top_label, iter, tmp;
-  enum machine_mode iter_mode;
+  enum machine_mode iter_mode = counter_mode (count);
   rtx piece_size = GEN_INT (GET_MODE_SIZE (mode) * unroll);
   rtx piece_size_mask = GEN_INT (~((GET_MODE_SIZE (mode) * unroll) - 1));
   rtx size;
@@ -13338,10 +13353,6 @@ expand_set_or_movmem_via_loop (rtx destmem, rtx srcmem,
   rtx y_addr;
   int i;
 
-  iter_mode = GET_MODE (count);
-  if (iter_mode == VOIDmode)
-    iter_mode = word_mode;
-
   top_label = gen_label_rtx ();
   out_label = gen_label_rtx ();
   iter = gen_reg_rtx (iter_mode);
@@ -13555,8 +13566,8 @@ expand_movmem_epilogue (rtx destmem, rtx srcmem,
            emit_strmov (destmem, srcmem, destptr, srcptr, DImode, offset);
          else
            {
-             emit_strmov (destmem, srcmem, destptr, srcptr, DImode, offset);
-             emit_strmov (destmem, srcmem, destptr, srcptr, DImode, offset + 4);
+             emit_strmov (destmem, srcmem, destptr, srcptr, SImode, offset);
+             emit_strmov (destmem, srcmem, destptr, srcptr, SImode, offset + 4);
            }
          offset += 8;
        }
@@ -13673,8 +13684,8 @@ expand_setmem_epilogue_via_loop (rtx destmem, rtx destptr, rtx value,
                                 rtx count, int max_size)
 {
   count =
-    expand_simple_binop (GET_MODE (count), AND, count, GEN_INT (max_size - 1),
-                        count, 1, OPTAB_DIRECT);
+    expand_simple_binop (counter_mode (count), AND, count,
+                        GEN_INT (max_size - 1), count, 1, OPTAB_DIRECT);
   expand_set_or_movmem_via_loop (destmem, NULL, destptr, NULL,
                                 gen_lowpart (QImode, value), count, QImode,
                                 1, max_size / 2);
@@ -14134,11 +14145,9 @@ ix86_expand_movmem (rtx dst, rtx src, rtx count_exp, rtx align_exp,
   gcc_assert (desired_align >= 1 && align >= 1);
 
   /* Ensure that alignment prologue won't copy past end of block.  */
-  if ((size_needed > 1 || (desired_align > 1 && desired_align > align))
-      && !count)
+  if (size_needed > 1 || (desired_align > 1 && desired_align > align))
     {
       epilogue_size_needed = MAX (size_needed - 1, desired_align - align);
-
       /* Epilogue always copies COUNT_EXP & EPILOGUE_SIZE_NEEDED bytes.
         Make sure it is power of 2.  */
       epilogue_size_needed = smallest_pow2_greater_than (epilogue_size_needed);
@@ -14146,8 +14155,10 @@ ix86_expand_movmem (rtx dst, rtx src, rtx count_exp, rtx align_exp,
       label = gen_label_rtx ();
       emit_cmp_and_jump_insns (count_exp,
                               GEN_INT (epilogue_size_needed),
-                              LTU, 0, GET_MODE (count_exp), 1, label);
-      if (expected_size == -1 || expected_size < epilogue_size_needed)
+                              LTU, 0, counter_mode (count_exp), 1, label);
+      if (GET_CODE (count_exp) == CONST_INT)
+       ;
+      else if (expected_size == -1 || expected_size < epilogue_size_needed)
        predict_jump (REG_BR_PROB_BASE * 60 / 100);
       else
        predict_jump (REG_BR_PROB_BASE * 20 / 100);
@@ -14247,7 +14258,7 @@ ix86_expand_movmem (rtx dst, rtx src, rtx count_exp, rtx align_exp,
       if (size_needed < epilogue_size_needed)
        {
          tmp =
-           expand_simple_binop (GET_MODE (count_exp), AND, count_exp,
+           expand_simple_binop (counter_mode (count_exp), AND, count_exp,
                                 GEN_INT (size_needed - 1), count_exp, 1,
                                 OPTAB_DIRECT);
          if (tmp != count_exp)
@@ -14403,7 +14414,7 @@ ix86_expand_setmem (rtx dst, rtx count_exp, rtx val_exp, rtx align_exp,
     return 0;
   gcc_assert (alg != no_stringop);
   if (!count)
-    count_exp = copy_to_mode_reg (GET_MODE (count_exp), count_exp);
+    count_exp = copy_to_mode_reg (counter_mode (count_exp), count_exp);
   destreg = copy_to_mode_reg (Pmode, XEXP (dst, 0));
   switch (alg)
     {
@@ -14446,11 +14457,9 @@ ix86_expand_setmem (rtx dst, rtx count_exp, rtx val_exp, rtx align_exp,
     promoted_val = promote_duplicated_reg_to_size (val_exp, size_needed,
                                                   desired_align, align);
   /* Ensure that alignment prologue won't copy past end of block.  */
-  if ((size_needed > 1 || (desired_align > 1 && desired_align > align))
-      && !count)
+  if (size_needed > 1 || (desired_align > 1 && desired_align > align))
     {
       epilogue_size_needed = MAX (size_needed - 1, desired_align - align);
-
       /* Epilogue always copies COUNT_EXP & EPILOGUE_SIZE_NEEDED bytes.
         Make sure it is power of 2.  */
       epilogue_size_needed = smallest_pow2_greater_than (epilogue_size_needed);
@@ -14464,8 +14473,10 @@ ix86_expand_setmem (rtx dst, rtx count_exp, rtx val_exp, rtx align_exp,
       label = gen_label_rtx ();
       emit_cmp_and_jump_insns (count_exp,
                               GEN_INT (epilogue_size_needed),
-                              LTU, 0, GET_MODE (count_exp), 1, label);
-      if (expected_size == -1 || expected_size <= epilogue_size_needed)
+                              LTU, 0, counter_mode (count_exp), 1, label);
+      if (GET_CODE (count_exp) == CONST_INT)
+       ;
+      else if (expected_size == -1 || expected_size <= epilogue_size_needed)
        predict_jump (REG_BR_PROB_BASE * 60 / 100);
       else
        predict_jump (REG_BR_PROB_BASE * 20 / 100);
@@ -14475,7 +14486,7 @@ ix86_expand_setmem (rtx dst, rtx count_exp, rtx val_exp, rtx align_exp,
       rtx hot_label = gen_label_rtx ();
       jump_around_label = gen_label_rtx ();
       emit_cmp_and_jump_insns (count_exp, GEN_INT (dynamic_check - 1),
-                              LEU, 0, GET_MODE (count_exp), 1, hot_label);
+                              LEU, 0, counter_mode (count_exp), 1, hot_label);
       predict_jump (REG_BR_PROB_BASE * 90 / 100);
       set_storage_via_libcall (dst, count_exp, val_exp, false);
       emit_jump (jump_around_label);
@@ -14558,7 +14569,7 @@ ix86_expand_setmem (rtx dst, rtx count_exp, rtx val_exp, rtx align_exp,
       if (size_needed < desired_align - align)
        {
          tmp =
-           expand_simple_binop (GET_MODE (count_exp), AND, count_exp,
+           expand_simple_binop (counter_mode (count_exp), AND, count_exp,
                                 GEN_INT (size_needed - 1), count_exp, 1,
                                 OPTAB_DIRECT);
          size_needed = desired_align - align + 1;

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 2e9bce6b315d2b85d5d3893e3589df5fe410a77e..c2e1df80a567c0e243ff41afa973237910491b89 100644 (file)
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,3 +1,7 @@
+2007-02-24  Jan Hubicka  <jh@suse.cz>
+
+       * gcc.c-torture/execute/pr30778.c: New testcase.
+
 2007-02-24  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 
        PR libgfortran/30918

diff --git a/gcc/testsuite/gcc.c-torture/execute/pr30778.c b/gcc/testsuite/gcc.c-torture/execute/pr30778.c
new file mode 100644 (file)
index 0000000..1b05189
--- /dev/null
+++ b/gcc/testsuite/gcc.c-torture/execute/pr30778.c
@@ -0,0 +1,34 @@
+extern void *memset (void *, int, unsigned long);
+extern void abort (void);
+
+struct reg_stat {
+  void *last_death;
+  void *last_set;
+  void *last_set_value;
+  int   last_set_label;
+  char  last_set_sign_bit_copies;
+  int   last_set_mode : 8;
+  char  last_set_invalid;
+  char sign_bit_copies;
+  long nonzero_bits;
+};
+
+static struct reg_stat *reg_stat;
+
+void __attribute__((noinline))
+init_reg_last (void)
+{
+  memset (reg_stat, 0, __builtin_offsetof (struct reg_stat, sign_bit_copies));
+}
+
+int main (void)
+{
+  struct reg_stat r;
+
+  reg_stat = &r;
+  r.nonzero_bits = -1;
+  init_reg_last ();
+  if (r.nonzero_bits != -1)
+    abort ();
+  return 0;
+}