PR70674: S/390: Add memory barrier to stack pointer restore from fpr.

This patches fixes a problem with stack variable accesses being
scheduled after the stack pointer restore instructions.  In the
testcase this happened with the stack variable 'a' accessed through the
frame pointer.

The existing stack_tie we have in the backend is basically useless
when trying to block stack variable accesses from being scheduled
across an insn.  The alias set of stack variables and the frame alias
set usually differ and hence aren't in conflict with each other.  The
solution appears to be a magic MEM term with a scratch register which
is handled as a full memory barrier when analyzing scheduling
dependencies.

With the patch a (clobber (mem:BLK (scratch))) is being added to the
restore instruction in order to prevent any memory operations to be
scheduled across the insn.  The patch does that only for the one case
where the stack pointer is restored from an FPR.  Theoretically this
might happen also in the case where the stack pointer gets restored
using a load multiple.  However, triggering that problem with
load-multiple appears to be much harder since the load-multiple will
restore the frame pointer as well.  So in order to see the problem a
different call-clobbered register would need to be used as temporary
stack pointer.

Another case which needs to be handled some day is the stack pointer
allocation part.  It needs to be a memory barrier as well.

Bootstrapped and regression tested with --with-arch z196 and z13 on
s390 and s390x.

-Andreas-

gcc/ChangeLog:

2016-04-20  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

	PR target/70674
	* config/s390/s390.c (s390_restore_gprs_from_fprs): Pick the new
	stack_restore_from_fpr pattern when restoring r15.
	(s390_optimize_prologue): Strip away the memory barrier in the
	parallel when trying to get rid of restore insns.
	* config/s390/s390.md ("stack_restore_from_fpr"): New insn
	definition for loading the stack pointer from an FPR.  Compared to
	the normal move insn this pattern includes a full memory barrier.

gcc/testsuite/ChangeLog:

2016-04-20  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>

	PR target/70674
	* gcc.target/s390/pr70674.c: New test.

From-SVN: r235234

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 30da178049a5b87b83c4850d9a97db300265a88b..685aba31ccb66dc217b92e241ec19e2e92651913 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,14 @@
+2016-04-20  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>
+
+       PR target/70674
+       * config/s390/s390.c (s390_restore_gprs_from_fprs): Pick the new
+       stack_restore_from_fpr pattern when restoring r15.
+       (s390_optimize_prologue): Strip away the memory barrier in the
+       parallel when trying to get rid of restore insns.
+       * config/s390/s390.md ("stack_restore_from_fpr"): New insn
+       definition for loading the stack pointer from an FPR.  Compared to
+       the normal move insn this pattern includes a full memory barrier.
+
 2016-04-19  Jakub Jelinek  <jakub@redhat.com>
 
        PR middle-end/70680

diff --git a/gcc/config/s390/s390.c b/gcc/config/s390/s390.c
index 1134d0f1e9dcbac6d566a51ca9f20320f9b2b686..a1d0930c071383f65905b2197f7673c02de415f7 100644 (file)
--- a/gcc/config/s390/s390.c
+++ b/gcc/config/s390/s390.c
@@ -10538,19 +10538,25 @@ s390_restore_gprs_from_fprs (void)
 
   for (i = 6; i < 16; i++)
     {
-      if (FP_REGNO_P (cfun_gpr_save_slot (i)))
-       {
-         rtx_insn *insn =
-           emit_move_insn (gen_rtx_REG (DImode, i),
-                           gen_rtx_REG (DImode, cfun_gpr_save_slot (i)));
-         df_set_regs_ever_live (i, true);
-         add_reg_note (insn, REG_CFA_RESTORE, gen_rtx_REG (DImode, i));
-         if (i == STACK_POINTER_REGNUM)
-           add_reg_note (insn, REG_CFA_DEF_CFA,
-                         plus_constant (Pmode, stack_pointer_rtx,
-                                        STACK_POINTER_OFFSET));
-         RTX_FRAME_RELATED_P (insn) = 1;
-       }
+      rtx_insn *insn;
+
+      if (!FP_REGNO_P (cfun_gpr_save_slot (i)))
+       continue;
+
+      rtx fpr = gen_rtx_REG (DImode, cfun_gpr_save_slot (i));
+
+      if (i == STACK_POINTER_REGNUM)
+       insn = emit_insn (gen_stack_restore_from_fpr (fpr));
+      else
+       insn = emit_move_insn (gen_rtx_REG (DImode, i), fpr);
+
+      df_set_regs_ever_live (i, true);
+      add_reg_note (insn, REG_CFA_RESTORE, gen_rtx_REG (DImode, i));
+      if (i == STACK_POINTER_REGNUM)
+       add_reg_note (insn, REG_CFA_DEF_CFA,
+                     plus_constant (Pmode, stack_pointer_rtx,
+                                    STACK_POINTER_OFFSET));
+      RTX_FRAME_RELATED_P (insn) = 1;
     }
 }
 
@@ -13032,37 +13038,46 @@ s390_optimize_prologue (void)
 
       /* Remove ldgr/lgdr instructions used for saving and restore
         GPRs if possible.  */
-      if (TARGET_Z10
-         && GET_CODE (pat) == SET
-         && GET_MODE (SET_SRC (pat)) == DImode
-         && REG_P (SET_SRC (pat))
-         && REG_P (SET_DEST (pat)))
+      if (TARGET_Z10)
        {
-         int src_regno = REGNO (SET_SRC (pat));
-         int dest_regno = REGNO (SET_DEST (pat));
-         int gpr_regno;
-         int fpr_regno;
+         rtx tmp_pat = pat;
 
-         if (!((GENERAL_REGNO_P (src_regno) && FP_REGNO_P (dest_regno))
-               || (FP_REGNO_P (src_regno) && GENERAL_REGNO_P (dest_regno))))
-           continue;
+         if (INSN_CODE (insn) == CODE_FOR_stack_restore_from_fpr)
+           tmp_pat = XVECEXP (pat, 0, 0);
 
-         gpr_regno = GENERAL_REGNO_P (src_regno) ? src_regno : dest_regno;
-         fpr_regno = FP_REGNO_P (src_regno) ? src_regno : dest_regno;
+         if (GET_CODE (tmp_pat) == SET
+             && GET_MODE (SET_SRC (tmp_pat)) == DImode
+             && REG_P (SET_SRC (tmp_pat))
+             && REG_P (SET_DEST (tmp_pat)))
+           {
+             int src_regno = REGNO (SET_SRC (tmp_pat));
+             int dest_regno = REGNO (SET_DEST (tmp_pat));
+             int gpr_regno;
+             int fpr_regno;
+
+             if (!((GENERAL_REGNO_P (src_regno)
+                    && FP_REGNO_P (dest_regno))
+                   || (FP_REGNO_P (src_regno)
+                       && GENERAL_REGNO_P (dest_regno))))
+               continue;
 
-         /* GPR must be call-saved, FPR must be call-clobbered.  */
-         if (!call_really_used_regs[fpr_regno]
-             || call_really_used_regs[gpr_regno])
-           continue;
+             gpr_regno = GENERAL_REGNO_P (src_regno) ? src_regno : dest_regno;
+             fpr_regno = FP_REGNO_P (src_regno) ? src_regno : dest_regno;
 
-         /* It must not happen that what we once saved in an FPR now
-            needs a stack slot.  */
-         gcc_assert (cfun_gpr_save_slot (gpr_regno) != SAVE_SLOT_STACK);
+             /* GPR must be call-saved, FPR must be call-clobbered.  */
+             if (!call_really_used_regs[fpr_regno]
+                 || call_really_used_regs[gpr_regno])
+               continue;
 
-         if (cfun_gpr_save_slot (gpr_regno) == SAVE_SLOT_NONE)
-           {
-             remove_insn (insn);
-             continue;
+             /* It must not happen that what we once saved in an FPR now
+                needs a stack slot.  */
+             gcc_assert (cfun_gpr_save_slot (gpr_regno) != SAVE_SLOT_STACK);
+
+             if (cfun_gpr_save_slot (gpr_regno) == SAVE_SLOT_NONE)
+               {
+                 remove_insn (insn);
+                 continue;
+               }
            }
        }
 

diff --git a/gcc/config/s390/s390.md b/gcc/config/s390/s390.md
index 5a9f1c8a6066d0211b904e6937f72c0bfa182590..2b4e8f4b2b2338bb5294312e7ec64d29a356149d 100644 (file)
--- a/gcc/config/s390/s390.md
+++ b/gcc/config/s390/s390.md
@@ -299,6 +299,8 @@
    (BASE_REGNUM                        13)
    ; Return address register.
    (RETURN_REGNUM              14)
+   ; Stack pointer register.
+   (STACK_REGNUM               15)
    ; Condition code register.
    (CC_REGNUM                  33)
    ; Thread local storage pointer register.
@@ -10387,6 +10389,14 @@
   [(set_attr "length" "0")])
 
 
+(define_insn "stack_restore_from_fpr"
+  [(set (reg:DI STACK_REGNUM)
+       (match_operand:DI 0 "register_operand" "f"))
+   (clobber (mem:BLK (scratch)))]
+  "TARGET_Z10"
+  "lgdr\t%%r15,%0"
+  [(set_attr "op_type"  "RRE")])
+
 ;
 ; Data prefetch patterns
 ;

diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index ed50f14795c5b2f0049615d183d3bf932b379f51..55e7bbda21f4394abf9f7fe37da769e869802820 100644 (file)
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,3 +1,8 @@
+2016-04-20  Andreas Krebbel  <krebbel@linux.vnet.ibm.com>
+
+       PR target/70674
+       * gcc.target/s390/pr70674.c: New test.
+
 2016-04-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 
        PR libgfortran/70684

diff --git a/gcc/testsuite/gcc.target/s390/pr70674.c b/gcc/testsuite/gcc.target/s390/pr70674.c
new file mode 100644 (file)
index 0000000..13bf271
--- /dev/null
+++ b/gcc/testsuite/gcc.target/s390/pr70674.c
@@ -0,0 +1,13 @@
+/* Test case for PR/70674.  */
+
+/* { dg-do compile { target s390x-*-* } } */
+/* { dg-options "-march=z10 -mtune=z196 -O2 -fno-omit-frame-pointer -fno-asynchronous-unwind-tables" } */
+
+void
+foo (void)
+{
+  volatile int a = 5;
+  (void) a;
+}
+
+/* { dg-final { scan-assembler-not "^.*lgdr.*\n.*\\(%r11\\)" } } */