diff options
Diffstat (limited to 'gcc/lra-constraints.c')
| -rw-r--r-- | gcc/lra-constraints.c | 18 |
1 files changed, 12 insertions, 6 deletions
diff --git a/gcc/lra-constraints.c b/gcc/lra-constraints.c index ddb91dd49cf..cdb2695acfa 100644 --- a/gcc/lra-constraints.c +++ b/gcc/lra-constraints.c @@ -928,10 +928,12 @@ match_reload (signed char out, signed char *ins, enum reg_class goal_class, they live in the same place. When we create a pseudo we assign value of original pseudo (if any) from which we created the new pseudo. If we create the pseudo from the - input pseudo, the new pseudo will no conflict with the input - pseudo which is wrong when the input pseudo lives after the - insn and as the new pseudo value is changed by the insn - output. Therefore we create the new pseudo from the output. + input pseudo, the new pseudo will have no conflict with the + input pseudo which is wrong when the input pseudo lives after + the insn and as the new pseudo value is changed by the insn + output. Therefore we create the new pseudo from the output + except the case when we have single matched dying input + pseudo. We cannot reuse the current output register because we might have a situation like "a <- a op b", where the constraints @@ -940,8 +942,12 @@ match_reload (signed char out, signed char *ins, enum reg_class goal_class, so that it doesn't clobber the current value of "a". */ new_in_reg = new_out_reg - = lra_create_new_reg_with_unique_value (outmode, out_rtx, - goal_class, ""); + = (ins[1] < 0 && REG_P (in_rtx) + && (int) REGNO (in_rtx) < lra_new_regno_start + && find_regno_note (curr_insn, REG_DEAD, REGNO (in_rtx)) + ? lra_create_new_reg (inmode, in_rtx, goal_class, "") + : lra_create_new_reg_with_unique_value (outmode, out_rtx, + goal_class, "")); } /* In operand can be got from transformations before processing insn constraints. One example of such transformations is subreg |