@@ -1090,16 +1090,22 @@ get_pressure_class_and_nregs (rtx insn, int *nregs)
}
/* Calculates cost and number of registers needed for moving invariant INV
- out of the loop and stores them to *COST and *REGS_NEEDED. */
+ out of the loop and stores them to *COST and *REGS_NEEDED. *CL will be
+ the REG_CLASS of INV. Return
+ -1: if INV is invalid.
+ 0: if INV and its depends_on have same reg_class
+ 1: if INV and its depends_on have different reg_classes. */
-static void
-get_inv_cost (struct invariant *inv, int *comp_cost, unsigned *regs_needed)
+static int
+get_inv_cost (struct invariant *inv, int *comp_cost, unsigned *regs_needed,
+ enum reg_class *cl)
{
int i, acomp_cost;
unsigned aregs_needed[N_REG_CLASSES];
unsigned depno;
struct invariant *dep;
bitmap_iterator bi;
+ int ret = 1;
/* Find the representative of the class of the equivalent invariants. */
inv = invariants[inv->eqto];
@@ -1115,7 +1121,7 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
if (inv->move
|| inv->stamp == actual_stamp)
- return;
+ return -1;
inv->stamp = actual_stamp;
if (! flag_ira_loop_pressure)
@@ -1127,6 +1133,8 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
pressure_class = get_pressure_class_and_nregs (inv->insn, &nregs);
regs_needed[pressure_class] += nregs;
+ *cl = pressure_class;
+ ret = 0;
}
if (!inv->cheap_address
@@ -1167,6 +1175,8 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
EXECUTE_IF_SET_IN_BITMAP (inv->depends_on, 0, depno, bi)
{
bool check_p;
+ enum reg_class dep_cl = ALL_REGS;
+ int dep_ret;
dep = invariants[depno];
@@ -1174,7 +1184,7 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
if (dep->move)
continue;
- get_inv_cost (dep, &acomp_cost, aregs_needed);
+ dep_ret = get_inv_cost (dep, &acomp_cost, aregs_needed, &dep_cl);
if (! flag_ira_loop_pressure)
check_p = aregs_needed[0] != 0;
@@ -1184,6 +1194,12 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
if (aregs_needed[ira_pressure_classes[i]] != 0)
break;
check_p = i < ira_pressure_classes_num;
+
+ if ((dep_ret == 1) || ((dep_ret == 0) && (*cl != dep_cl)))
+ {
+ *cl = ALL_REGS;
+ ret = 1;
+ }
}
if (check_p
/* We need to check always_executed, since if the original value of
@@ -1217,6 +1233,7 @@ get_inv_cost (struct invariant *inv, int
*comp_cost, unsigned *regs_needed)
}
(*comp_cost) += acomp_cost;
}
+ return ret;
}
/* Calculates gain for eliminating invariant INV. REGS_USED is the number
@@ -1231,10 +1248,12 @@ gain_for_invariant (struct invariant *inv,
unsigned *regs_needed,
bool speed, bool call_p)
{
int comp_cost, size_cost;
+ enum reg_class cl;
+ int ret;
actual_stamp++;
- get_inv_cost (inv, &comp_cost, regs_needed);
+ ret = get_inv_cost (inv, &comp_cost, regs_needed, &cl);
if (! flag_ira_loop_pressure)
{
@@ -1243,6 +1262,11 @@ gain_for_invariant (struct invariant *inv,
unsigned *regs_needed,
- estimate_reg_pressure_cost (new_regs[0],
regs_used, speed, call_p));
}
+ else if (ret < 0)
+ return -1;
+ else if ((ret == 0) && (cl == NO_REGS))
+ /* Hoist it anyway since it does not impact register pressure. */
+ return 1;
else
{
int i;
@@ -1251,6 +1275,10 @@ gain_for_invariant (struct invariant *inv,
unsigned *regs_needed,
for (i = 0; i < ira_pressure_classes_num; i++)
{
pressure_class = ira_pressure_classes[i];
+
+ if (!reg_classes_intersect_p (pressure_class, cl))
+ continue;
+
if ((int) new_regs[pressure_class]
+ (int) regs_needed[pressure_class]
+ LOOP_DATA (curr_loop)->max_reg_pressure[pressure_class]
b/gcc/testsuite/gcc.dg/ira-loop-pressure.c
new file mode 100644
@@ -0,0 +1,32 @@
+/* Loop2_invariants pass should distinguish register pressures of different
+ register classes. In this case, register pressue of INT is high. But
+ we can still move the FP invariant out of the loop. */
+
+/* { dg-do compile { target { x86_64-*-* && lp64 } } } */
+/* { dg-options "-O2 -fira-loop-pressure -fdump-rtl-loop2_invariant " } */
+
+float tt;
+extern void foo2 (int *, int *, int *, int *, int *, int *);
+extern int foo3 (int, int, int, int, int, int);
+int foo (int a, int b, int c, int d)
+{
+ int i = a;
+ int t1, t2, t3, t4, t5, t6;
+ t1 = t2 = t3 = t4 = t5 = t6 = 0;
+
+ for (; i > 0; i += c)
+ {
+ tt += 123456.0;
+ if (d > t1 + t2)
+ tt = 3.0;
+ foo2 (&t1, &t2, &t3, &t4, &t5, &t6);
+ t1 += t3 + t4 + a + b;
+ t2 -= t5 - t6 - c - d;
+ }
+
+ return foo3 (t1, t2, t3, t4, t5, t6);
+}
+
+/* { dg-final { scan-rtl-dump "Decided to move invariant"
"loop2_invariant" } } */
+/* { dg-final { cleanup-rtl-dump "loop2_invariant" } } */