diff options
Diffstat (limited to 'gcc/config/i386/i386.cc')
| -rw-r--r-- | gcc/config/i386/i386.cc | 48 |
1 files changed, 43 insertions, 5 deletions
diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc index 587b2bd..75a9cb6 100644 --- a/gcc/config/i386/i386.cc +++ b/gcc/config/i386/i386.cc @@ -598,6 +598,20 @@ ix86_canonicalize_comparison (int *code, rtx *op0, rtx *op1, } } + /* SUB (a, b) underflows precisely when a < b. Convert + (compare (minus (a b)) a) to (compare (a b)) + to match *sub<mode>_3 pattern. */ + if (!op0_preserve_value + && (*code == GTU || *code == LEU) + && GET_CODE (*op0) == MINUS + && rtx_equal_p (XEXP (*op0, 0), *op1)) + { + *op1 = XEXP (*op0, 1); + *op0 = XEXP (*op0, 0); + *code = (int) swap_condition ((enum rtx_code) *code); + return; + } + /* Swap operands of GTU comparison to canonicalize addcarry/subborrow comparison. */ if (!op0_preserve_value @@ -23753,9 +23767,15 @@ x86_print_call_or_nop (FILE *file, const char *target, const char *label) { if (flag_nop_mcount || !strcmp (target, "nop")) - /* 5 byte nop: nopl 0(%[re]ax,%[re]ax,1) */ - fprintf (file, "%s" ASM_BYTE "0x0f, 0x1f, 0x44, 0x00, 0x00\n", - label); + { + if (TARGET_16BIT) + /* 3 byte no-op: lea 0(%si), %si */ + fprintf (file, "%s" ASM_BYTE "0x8d, 0x74, 0x00\n", label); + else + /* 5 byte nop: nopl 0(%[re]ax,%[re]ax,1) */ + fprintf (file, "%s" ASM_BYTE "0x0f, 0x1f, 0x44, 0x00, 0x00\n", + label); + } else if (!TARGET_PECOFF && flag_pic) { gcc_assert (flag_plt); @@ -25089,7 +25109,7 @@ i386_solaris_elf_named_section (const char *name, unsigned int flags, return; } -#ifndef USE_GAS +#if !HAVE_GNU_AS if (HAVE_COMDAT_GROUP && flags & SECTION_LINKONCE) { solaris_elf_asm_comdat_section (name, flags, decl); @@ -26377,7 +26397,20 @@ ix86_vector_costs::add_stmt_cost (int count, vect_cost_for_stmt kind, (TREE_OPERAND (gimple_assign_rhs1 (def), 0)))))) { if (fp) - m_num_sse_needed[where]++; + { + /* Scalar FP values residing in x87 registers need to be + spilled and reloaded. */ + auto mode2 = TYPE_MODE (TREE_TYPE (op)); + if (IS_STACK_MODE (mode2)) + { + int cost + = (ix86_cost->hard_register.fp_store[mode2 == SFmode + ? 0 : 1] + + ix86_cost->sse_load[sse_store_index (mode2)]); + stmt_cost += COSTS_N_INSNS (cost) / 2; + } + m_num_sse_needed[where]++; + } else { m_num_gpr_needed[where]++; @@ -26595,6 +26628,11 @@ ix86_vector_costs::finish_cost (const vector_costs *scalar_costs) if (loop_vinfo && !LOOP_VINFO_EPILOGUE_P (loop_vinfo) && LOOP_VINFO_VECT_FACTOR (loop_vinfo).to_constant () > 2 + /* Avoid a masked epilog if cascaded epilogues eventually get us + to one with VF 1 as that means no scalar epilog at all. */ + && !((GET_MODE_SIZE (loop_vinfo->vector_mode) + / LOOP_VINFO_VECT_FACTOR (loop_vinfo).to_constant () == 16) + && ix86_tune_features[X86_TUNE_AVX512_TWO_EPILOGUES]) && ix86_tune_features[X86_TUNE_AVX512_MASKED_EPILOGUES] && !OPTION_SET_P (param_vect_partial_vector_usage)) { |