+2006-10-22 H.J. Lu <hongjiu.lu@intel.com>
+
+ * config.gcc (i[34567]86-*-*): Add tmmintrin.h to extra_headers.
+ (x86_64-*-*): Likewise.
+
+ * config/i386/i386.c (pta_flags): Add PTA_SSSE3.
+ (override_options): Check SSSE3.
+ (ix86_builtins): Add IX86_BUILTIN_PHADDW, IX86_BUILTIN_PHADDD,
+ IX86_BUILTIN_PHADDSW, IX86_BUILTIN_PHSUBW, IX86_BUILTIN_PHSUBD,
+ IX86_BUILTIN_PHSUBSW, IX86_BUILTIN_PMADDUBSW,
+ IX86_BUILTIN_PMULHRSW, IX86_BUILTIN_PSHUFB,
+ IX86_BUILTIN_PSIGNB, IX86_BUILTIN_PSIGNW, IX86_BUILTIN_PSIGND,
+ IX86_BUILTIN_PALIGNR, IX86_BUILTIN_PABSB, IX86_BUILTIN_PABSW,
+ IX86_BUILTIN_PABSD, IX86_BUILTIN_PHADDW128,
+ IX86_BUILTIN_PHADDD128, IX86_BUILTIN_PHADDSW128,
+ IX86_BUILTIN_PHSUBW128, IX86_BUILTIN_PHSUBD128,
+ IX86_BUILTIN_PHSUBSW128, IX86_BUILTIN_PMADDUBSW128,
+ IX86_BUILTIN_PMULHRSW128, IX86_BUILTIN_PSHUFB128,
+ IX86_BUILTIN_PSIGNB128, IX86_BUILTIN_PSIGNW128,
+ IX86_BUILTIN_PSIGND128, IX86_BUILTIN_PALIGNR128,
+ IX86_BUILTIN_PABSB128, IX86_BUILTIN_PABSW128 and
+ IX86_BUILTIN_PABSD128.
+ (bdesc_2arg): Add SSSE3.
+ (bdesc_1arg): Likewise.
+ (ix86_init_mmx_sse_builtins): Support SSSE3.
+ (ix86_expand_builtin): Likewise.
+ * config/i386/i386.h (TARGET_CPU_CPP_BUILTINS): Likewise.
+
+ * config/i386/i386.md (UNSPEC_PSHUFB): New.
+ (UNSPEC_PSIGN): Likewise.
+ (UNSPEC_PALIGNR): Likewise.
+ Include mmx.md before sse.md.
+
+ * config/i386/i386.opt: Add -mssse3.
+
+ * config/i386/sse.md (ssse3_phaddwv8hi3): New pattern for SSSE3.
+ (ssse3_phaddwv4hi3): Likewise.
+ (ssse3_phadddv4si3): Likewise.
+ (ssse3_phadddv2si3): Likewise.
+ (ssse3_phaddswv8hi3): Likewise.
+ (ssse3_phaddswv4hi3): Likewise.
+ (ssse3_phsubwv8hi3): Likewise.
+ (ssse3_phsubwv4hi3): Likewise.
+ (ssse3_phsubdv4si3): Likewise.
+ (ssse3_phsubdv2si3): Likewise.
+ (ssse3_phsubswv8hi3): Likewise.
+ (ssse3_phsubswv4hi3): Likewise.
+ (ssse3_pmaddubswv8hi3): Likewise.
+ (ssse3_pmaddubswv4hi3): Likewise.
+ (ssse3_pmulhrswv8hi3): Likewise.
+ (ssse3_pmulhrswv4hi3): Likewise.
+ (ssse3_pshufbv16qi3): Likewise.
+ (ssse3_pshufbv8qi3): Likewise.
+ (ssse3_psign<mode>3): Likewise.
+ (ssse3_psign<mode>3): Likewise.
+ (ssse3_palignrti): Likewise.
+ (ssse3_palignrdi): Likewise.
+ (abs<mode>2): Likewise.
+ (abs<mode>2): Likewise.
+
+ * config/i386/tmmintrin.h: New file.
+
+ * doc/extend.texi: Document SSSE3 built-in functions.
+
+ * doc/invoke.texi: Document -mssse3/-mno-ssse3 switches.
+
2006-10-22 Ira Rosen <irar@il.ibm.com>
* tree-vect-transform.c (vectorizable_load): Use the type of the
;;
i[34567]86-*-*)
cpu_type=i386
- extra_headers="mmintrin.h mm3dnow.h xmmintrin.h emmintrin.h pmmintrin.h"
+ extra_headers="mmintrin.h mm3dnow.h xmmintrin.h emmintrin.h
+ pmmintrin.h tmmintrin.h"
;;
x86_64-*-*)
cpu_type=i386
- extra_headers="mmintrin.h mm3dnow.h xmmintrin.h emmintrin.h pmmintrin.h"
+ extra_headers="mmintrin.h mm3dnow.h xmmintrin.h emmintrin.h
+ pmmintrin.h tmmintrin.h"
need_64bit_hwint=yes
;;
ia64-*-*)
PTA_PREFETCH_SSE = 16,
PTA_3DNOW = 32,
PTA_3DNOW_A = 64,
- PTA_64BIT = 128
+ PTA_64BIT = 128,
+ PTA_SSSE3 = 256
} flags;
}
const processor_alias_table[] =
if (processor_alias_table[i].flags & PTA_SSE3
&& !(target_flags_explicit & MASK_SSE3))
target_flags |= MASK_SSE3;
+ if (processor_alias_table[i].flags & PTA_SSSE3
+ && !(target_flags_explicit & MASK_SSSE3))
+ target_flags |= MASK_SSSE3;
if (processor_alias_table[i].flags & PTA_PREFETCH_SSE)
x86_prefetch_sse = true;
if (TARGET_64BIT && !(processor_alias_table[i].flags & PTA_64BIT))
if (!TARGET_80387)
target_flags |= MASK_NO_FANCY_MATH_387;
+ /* Turn on SSE3 builtins for -mssse3. */
+ if (TARGET_SSSE3)
+ target_flags |= MASK_SSE3;
+
/* Turn on SSE2 builtins for -msse3. */
if (TARGET_SSE3)
target_flags |= MASK_SSE2;
IX86_BUILTIN_MONITOR,
IX86_BUILTIN_MWAIT,
+ /* SSSE3. */
+ IX86_BUILTIN_PHADDW,
+ IX86_BUILTIN_PHADDD,
+ IX86_BUILTIN_PHADDSW,
+ IX86_BUILTIN_PHSUBW,
+ IX86_BUILTIN_PHSUBD,
+ IX86_BUILTIN_PHSUBSW,
+ IX86_BUILTIN_PMADDUBSW,
+ IX86_BUILTIN_PMULHRSW,
+ IX86_BUILTIN_PSHUFB,
+ IX86_BUILTIN_PSIGNB,
+ IX86_BUILTIN_PSIGNW,
+ IX86_BUILTIN_PSIGND,
+ IX86_BUILTIN_PALIGNR,
+ IX86_BUILTIN_PABSB,
+ IX86_BUILTIN_PABSW,
+ IX86_BUILTIN_PABSD,
+
+ IX86_BUILTIN_PHADDW128,
+ IX86_BUILTIN_PHADDD128,
+ IX86_BUILTIN_PHADDSW128,
+ IX86_BUILTIN_PHSUBW128,
+ IX86_BUILTIN_PHSUBD128,
+ IX86_BUILTIN_PHSUBSW128,
+ IX86_BUILTIN_PMADDUBSW128,
+ IX86_BUILTIN_PMULHRSW128,
+ IX86_BUILTIN_PSHUFB128,
+ IX86_BUILTIN_PSIGNB128,
+ IX86_BUILTIN_PSIGNW128,
+ IX86_BUILTIN_PSIGND128,
+ IX86_BUILTIN_PALIGNR128,
+ IX86_BUILTIN_PABSB128,
+ IX86_BUILTIN_PABSW128,
+ IX86_BUILTIN_PABSD128,
+
IX86_BUILTIN_VEC_INIT_V2SI,
IX86_BUILTIN_VEC_INIT_V4HI,
IX86_BUILTIN_VEC_INIT_V8QI,
{ MASK_SSE3, CODE_FOR_sse3_haddv4sf3, "__builtin_ia32_haddps", IX86_BUILTIN_HADDPS, 0, 0 },
{ MASK_SSE3, CODE_FOR_sse3_haddv2df3, "__builtin_ia32_haddpd", IX86_BUILTIN_HADDPD, 0, 0 },
{ MASK_SSE3, CODE_FOR_sse3_hsubv4sf3, "__builtin_ia32_hsubps", IX86_BUILTIN_HSUBPS, 0, 0 },
- { MASK_SSE3, CODE_FOR_sse3_hsubv2df3, "__builtin_ia32_hsubpd", IX86_BUILTIN_HSUBPD, 0, 0 }
+ { MASK_SSE3, CODE_FOR_sse3_hsubv2df3, "__builtin_ia32_hsubpd", IX86_BUILTIN_HSUBPD, 0, 0 },
+
+ /* SSSE3 */
+ { MASK_SSSE3, CODE_FOR_ssse3_phaddwv8hi3, "__builtin_ia32_phaddw128", IX86_BUILTIN_PHADDW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phaddwv4hi3, "__builtin_ia32_phaddw", IX86_BUILTIN_PHADDW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phadddv4si3, "__builtin_ia32_phaddd128", IX86_BUILTIN_PHADDD128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phadddv2si3, "__builtin_ia32_phaddd", IX86_BUILTIN_PHADDD, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phaddswv8hi3, "__builtin_ia32_phaddsw128", IX86_BUILTIN_PHADDSW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phaddswv4hi3, "__builtin_ia32_phaddsw", IX86_BUILTIN_PHADDSW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubwv8hi3, "__builtin_ia32_phsubw128", IX86_BUILTIN_PHSUBW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubwv4hi3, "__builtin_ia32_phsubw", IX86_BUILTIN_PHSUBW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubdv4si3, "__builtin_ia32_phsubd128", IX86_BUILTIN_PHSUBD128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubdv2si3, "__builtin_ia32_phsubd", IX86_BUILTIN_PHSUBD, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubswv8hi3, "__builtin_ia32_phsubsw128", IX86_BUILTIN_PHSUBSW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_phsubswv4hi3, "__builtin_ia32_phsubsw", IX86_BUILTIN_PHSUBSW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pmaddubswv8hi3, "__builtin_ia32_pmaddubsw128", IX86_BUILTIN_PMADDUBSW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pmaddubswv4hi3, "__builtin_ia32_pmaddubsw", IX86_BUILTIN_PMADDUBSW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pmulhrswv8hi3, "__builtin_ia32_pmulhrsw128", IX86_BUILTIN_PMULHRSW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pmulhrswv4hi3, "__builtin_ia32_pmulhrsw", IX86_BUILTIN_PMULHRSW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pshufbv16qi3, "__builtin_ia32_pshufb128", IX86_BUILTIN_PSHUFB128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_pshufbv8qi3, "__builtin_ia32_pshufb", IX86_BUILTIN_PSHUFB, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv16qi3, "__builtin_ia32_psignb128", IX86_BUILTIN_PSIGNB128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv8qi3, "__builtin_ia32_psignb", IX86_BUILTIN_PSIGNB, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv8hi3, "__builtin_ia32_psignw128", IX86_BUILTIN_PSIGNW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv4hi3, "__builtin_ia32_psignw", IX86_BUILTIN_PSIGNW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv4si3, "__builtin_ia32_psignd128", IX86_BUILTIN_PSIGND128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_ssse3_psignv2si3, "__builtin_ia32_psignd", IX86_BUILTIN_PSIGND, 0, 0 }
};
static const struct builtin_description bdesc_1arg[] =
/* SSE3 */
{ MASK_SSE3, CODE_FOR_sse3_movshdup, 0, IX86_BUILTIN_MOVSHDUP, 0, 0 },
{ MASK_SSE3, CODE_FOR_sse3_movsldup, 0, IX86_BUILTIN_MOVSLDUP, 0, 0 },
+
+ /* SSSE3 */
+ { MASK_SSSE3, CODE_FOR_absv16qi2, "__builtin_ia32_pabsb128", IX86_BUILTIN_PABSB128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_absv8qi2, "__builtin_ia32_pabsb", IX86_BUILTIN_PABSB, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_absv8hi2, "__builtin_ia32_pabsw128", IX86_BUILTIN_PABSW128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_absv4hi2, "__builtin_ia32_pabsw", IX86_BUILTIN_PABSW, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_absv4si2, "__builtin_ia32_pabsd128", IX86_BUILTIN_PABSD128, 0, 0 },
+ { MASK_SSSE3, CODE_FOR_absv2si2, "__builtin_ia32_pabsd", IX86_BUILTIN_PABSD, 0, 0 },
};
static void
/* Normal vector unops. */
tree v4sf_ftype_v4sf
= build_function_type_list (V4SF_type_node, V4SF_type_node, NULL_TREE);
+ tree v16qi_ftype_v16qi
+ = build_function_type_list (V16QI_type_node, V16QI_type_node, NULL_TREE);
+ tree v8hi_ftype_v8hi
+ = build_function_type_list (V8HI_type_node, V8HI_type_node, NULL_TREE);
+ tree v4si_ftype_v4si
+ = build_function_type_list (V4SI_type_node, V4SI_type_node, NULL_TREE);
+ tree v8qi_ftype_v8qi
+ = build_function_type_list (V8QI_type_node, V8QI_type_node, NULL_TREE);
+ tree v4hi_ftype_v4hi
+ = build_function_type_list (V4HI_type_node, V4HI_type_node, NULL_TREE);
/* Normal vector binops. */
tree v4sf_ftype_v4sf_v4sf
long_long_unsigned_type_node,
long_long_unsigned_type_node, NULL_TREE);
+ tree di_ftype_di_di_int
+ = build_function_type_list (long_long_unsigned_type_node,
+ long_long_unsigned_type_node,
+ long_long_unsigned_type_node,
+ integer_type_node, NULL_TREE);
+
tree v2si_ftype_v2sf
= build_function_type_list (V2SI_type_node, V2SF_type_node, NULL_TREE);
tree v2sf_ftype_v2si
tree v2di_ftype_v2di_int
= build_function_type_list (V2DI_type_node,
V2DI_type_node, integer_type_node, NULL_TREE);
+ tree v2di_ftype_v2di_v2di_int
+ = build_function_type_list (V2DI_type_node, V2DI_type_node,
+ V2DI_type_node, integer_type_node, NULL_TREE);
tree v4si_ftype_v4si_int
= build_function_type_list (V4SI_type_node,
V4SI_type_node, integer_type_node, NULL_TREE);
def_builtin (d->mask, d->name, type, d->code);
}
+ /* Add all builtins that are more or less simple operations on 1 operand. */
+ for (i = 0, d = bdesc_1arg; i < ARRAY_SIZE (bdesc_1arg); i++, d++)
+ {
+ enum machine_mode mode;
+ tree type;
+
+ if (d->name == 0)
+ continue;
+ mode = insn_data[d->icode].operand[1].mode;
+
+ switch (mode)
+ {
+ case V16QImode:
+ type = v16qi_ftype_v16qi;
+ break;
+ case V8HImode:
+ type = v8hi_ftype_v8hi;
+ break;
+ case V4SImode:
+ type = v4si_ftype_v4si;
+ break;
+ case V2DFmode:
+ type = v2df_ftype_v2df;
+ break;
+ case V4SFmode:
+ type = v4sf_ftype_v4sf;
+ break;
+ case V8QImode:
+ type = v8qi_ftype_v8qi;
+ break;
+ case V4HImode:
+ type = v4hi_ftype_v4hi;
+ break;
+ case V2SImode:
+ type = v2si_ftype_v2si;
+ break;
+
+ default:
+ abort ();
+ }
+
+ def_builtin (d->mask, d->name, type, d->code);
+ }
+
/* Add the remaining MMX insns with somewhat more complicated types. */
def_builtin (MASK_MMX, "__builtin_ia32_emms", void_ftype_void, IX86_BUILTIN_EMMS);
def_builtin (MASK_MMX, "__builtin_ia32_psllw", v4hi_ftype_v4hi_di, IX86_BUILTIN_PSLLW);
def_builtin (MASK_SSE3, "__builtin_ia32_lddqu",
v16qi_ftype_pcchar, IX86_BUILTIN_LDDQU);
+ /* SSSE3. */
+ def_builtin (MASK_SSSE3, "__builtin_ia32_palignr128",
+ v2di_ftype_v2di_v2di_int, IX86_BUILTIN_PALIGNR128);
+ def_builtin (MASK_SSSE3, "__builtin_ia32_palignr", di_ftype_di_di_int,
+ IX86_BUILTIN_PALIGNR);
+
/* Access to the vec_init patterns. */
ftype = build_function_type_list (V2SI_type_node, integer_type_node,
integer_type_node, NULL_TREE);
tree arglist = TREE_OPERAND (exp, 1);
tree arg0, arg1, arg2;
rtx op0, op1, op2, pat;
- enum machine_mode tmode, mode0, mode1, mode2;
+ enum machine_mode tmode, mode0, mode1, mode2, mode3;
unsigned int fcode = DECL_FUNCTION_CODE (fndecl);
switch (fcode)
return ix86_expand_unop_builtin (CODE_FOR_sse3_lddqu, arglist,
target, 1);
+ case IX86_BUILTIN_PALIGNR:
+ case IX86_BUILTIN_PALIGNR128:
+ if (fcode == IX86_BUILTIN_PALIGNR)
+ {
+ icode = CODE_FOR_ssse3_palignrdi;
+ mode = DImode;
+ }
+ else
+ {
+ icode = CODE_FOR_ssse3_palignrti;
+ mode = V2DImode;
+ }
+ arg0 = TREE_VALUE (arglist);
+ arg1 = TREE_VALUE (TREE_CHAIN (arglist));
+ arg2 = TREE_VALUE (TREE_CHAIN (TREE_CHAIN (arglist)));
+ op0 = expand_expr (arg0, NULL_RTX, VOIDmode, 0);
+ op1 = expand_expr (arg1, NULL_RTX, VOIDmode, 0);
+ op2 = expand_expr (arg2, NULL_RTX, VOIDmode, 0);
+ tmode = insn_data[icode].operand[0].mode;
+ mode1 = insn_data[icode].operand[1].mode;
+ mode2 = insn_data[icode].operand[2].mode;
+ mode3 = insn_data[icode].operand[3].mode;
+
+ if (! (*insn_data[icode].operand[1].predicate) (op0, mode1))
+ {
+ op0 = copy_to_reg (op0);
+ op0 = simplify_gen_subreg (mode1, op0, GET_MODE (op0), 0);
+ }
+ if (! (*insn_data[icode].operand[2].predicate) (op1, mode2))
+ {
+ op1 = copy_to_reg (op1);
+ op1 = simplify_gen_subreg (mode2, op1, GET_MODE (op1), 0);
+ }
+ if (! (*insn_data[icode].operand[3].predicate) (op2, mode3))
+ {
+ error ("shift must be an immediate");
+ return const0_rtx;
+ }
+ target = gen_reg_rtx (mode);
+ pat = GEN_FCN (icode) (simplify_gen_subreg (tmode, target, mode, 0),
+ op0, op1, op2);
+ if (! pat)
+ return 0;
+ emit_insn (pat);
+ return target;
+
case IX86_BUILTIN_VEC_INIT_V2SI:
case IX86_BUILTIN_VEC_INIT_V4HI:
case IX86_BUILTIN_VEC_INIT_V8QI:
builtin_define ("__SSE2__"); \
if (TARGET_SSE3) \
builtin_define ("__SSE3__"); \
+ if (TARGET_SSSE3) \
+ builtin_define ("__SSSE3__"); \
if (TARGET_SSE_MATH && TARGET_SSE) \
builtin_define ("__SSE_MATH__"); \
if (TARGET_SSE_MATH && TARGET_SSE2) \
(UNSPEC_SP_TEST 101)
(UNSPEC_SP_TLS_SET 102)
(UNSPEC_SP_TLS_TEST 103)
+
+ ; SSSE3
+ (UNSPEC_PSHUFB 120)
+ (UNSPEC_PSIGN 121)
+ (UNSPEC_PALIGNR 122)
])
(define_constants
}
[(set_attr "type" "multi")])
-(include "sse.md")
(include "mmx.md")
+(include "sse.md")
(include "sync.md")
Target Report Mask(SSE3)
Support MMX, SSE, SSE2 and SSE3 built-in functions and code generation
+mssse3
+Target Report Mask(SSSE3)
+Support MMX, SSE, SSE2, SSE3 and SSSE3 built-in functions and code generation
+
msseregparm
Target RejectNegative Mask(SSEREGPARM)
Use SSE register passing conventions for SF and DF mode
;; zero extended to 64bit, we only need to set up 32bit registers.
"monitor"
[(set_attr "length" "3")])
+
+;; SSSE3
+(define_insn "ssse3_phaddwv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (vec_concat:V8HI
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI
+ (match_operand:V8HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 5)])))
+ (plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 7)])))))
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI
+ (match_operand:V8HI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 5)])))
+ (plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 7)])))))))]
+ "TARGET_SSSE3"
+ "phaddw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phaddwv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI
+ (match_operand:V4HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (plus:HI
+ (vec_select:HI
+ (match_operand:V4HI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phaddw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_phadddv4si3"
+ [(set (match_operand:V4SI 0 "register_operand" "=x")
+ (vec_concat:V4SI
+ (vec_concat:V2SI
+ (plus:SI
+ (vec_select:SI
+ (match_operand:V4SI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 1)])))
+ (plus:SI
+ (vec_select:SI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2SI
+ (plus:SI
+ (vec_select:SI
+ (match_operand:V4SI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 1)])))
+ (plus:SI
+ (vec_select:SI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phaddd\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phadddv2si3"
+ [(set (match_operand:V2SI 0 "register_operand" "=y")
+ (vec_concat:V2SI
+ (plus:SI
+ (vec_select:SI
+ (match_operand:V2SI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 1)])))
+ (plus:SI
+ (vec_select:SI
+ (match_operand:V2SI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 1)])))))]
+ "TARGET_SSSE3"
+ "phaddd\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_phaddswv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (vec_concat:V8HI
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI
+ (match_operand:V8HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 5)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 7)])))))
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI
+ (match_operand:V8HI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 5)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 7)])))))))]
+ "TARGET_SSSE3"
+ "phaddsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phaddswv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI
+ (match_operand:V4HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_plus:HI
+ (vec_select:HI
+ (match_operand:V4HI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (ss_plus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phaddsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_phsubwv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (vec_concat:V8HI
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI
+ (match_operand:V8HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 5)])))
+ (minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 7)])))))
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI
+ (match_operand:V8HI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 5)])))
+ (minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 7)])))))))]
+ "TARGET_SSSE3"
+ "phsubw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phsubwv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI
+ (match_operand:V4HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (minus:HI
+ (vec_select:HI
+ (match_operand:V4HI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phsubw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_phsubdv4si3"
+ [(set (match_operand:V4SI 0 "register_operand" "=x")
+ (vec_concat:V4SI
+ (vec_concat:V2SI
+ (minus:SI
+ (vec_select:SI
+ (match_operand:V4SI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 1)])))
+ (minus:SI
+ (vec_select:SI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2SI
+ (minus:SI
+ (vec_select:SI
+ (match_operand:V4SI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 1)])))
+ (minus:SI
+ (vec_select:SI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phsubd\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phsubdv2si3"
+ [(set (match_operand:V2SI 0 "register_operand" "=y")
+ (vec_concat:V2SI
+ (minus:SI
+ (vec_select:SI
+ (match_operand:V2SI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 1) (parallel [(const_int 1)])))
+ (minus:SI
+ (vec_select:SI
+ (match_operand:V2SI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:SI (match_dup 2) (parallel [(const_int 1)])))))]
+ "TARGET_SSSE3"
+ "phsubd\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_phsubswv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (vec_concat:V8HI
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI
+ (match_operand:V8HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 5)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 7)])))))
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI
+ (match_operand:V8HI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 4)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 5)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 6)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 7)])))))))]
+ "TARGET_SSSE3"
+ "phsubsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_phsubswv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (vec_concat:V4HI
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI
+ (match_operand:V4HI 1 "register_operand" "0")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 1) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 1) (parallel [(const_int 3)]))))
+ (vec_concat:V2HI
+ (ss_minus:HI
+ (vec_select:HI
+ (match_operand:V4HI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
+ (ss_minus:HI
+ (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
+ (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))))]
+ "TARGET_SSSE3"
+ "phsubsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_pmaddubswv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (ss_plus:V8HI
+ (mult:V8HI
+ (zero_extend:V8HI
+ (vec_select:V4QI
+ (match_operand:V16QI 1 "nonimmediate_operand" "%0")
+ (parallel [(const_int 0)
+ (const_int 2)
+ (const_int 4)
+ (const_int 6)
+ (const_int 8)
+ (const_int 10)
+ (const_int 12)
+ (const_int 14)])))
+ (sign_extend:V8HI
+ (vec_select:V8QI
+ (match_operand:V16QI 2 "nonimmediate_operand" "xm")
+ (parallel [(const_int 0)
+ (const_int 2)
+ (const_int 4)
+ (const_int 6)
+ (const_int 8)
+ (const_int 10)
+ (const_int 12)
+ (const_int 14)]))))
+ (mult:V8HI
+ (zero_extend:V8HI
+ (vec_select:V16QI (match_dup 1)
+ (parallel [(const_int 1)
+ (const_int 3)
+ (const_int 5)
+ (const_int 7)
+ (const_int 9)
+ (const_int 11)
+ (const_int 13)
+ (const_int 15)])))
+ (sign_extend:V8HI
+ (vec_select:V16QI (match_dup 2)
+ (parallel [(const_int 1)
+ (const_int 3)
+ (const_int 5)
+ (const_int 7)
+ (const_int 9)
+ (const_int 11)
+ (const_int 13)
+ (const_int 15)]))))))]
+ "TARGET_SSSE3"
+ "pmaddubsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_pmaddubswv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (ss_plus:V4HI
+ (mult:V4HI
+ (zero_extend:V4HI
+ (vec_select:V4QI
+ (match_operand:V8QI 1 "nonimmediate_operand" "%0")
+ (parallel [(const_int 0)
+ (const_int 2)
+ (const_int 4)
+ (const_int 6)])))
+ (sign_extend:V4HI
+ (vec_select:V4QI
+ (match_operand:V8QI 2 "nonimmediate_operand" "ym")
+ (parallel [(const_int 0)
+ (const_int 2)
+ (const_int 4)
+ (const_int 6)]))))
+ (mult:V4HI
+ (zero_extend:V4HI
+ (vec_select:V8QI (match_dup 1)
+ (parallel [(const_int 1)
+ (const_int 3)
+ (const_int 5)
+ (const_int 7)])))
+ (sign_extend:V4HI
+ (vec_select:V8QI (match_dup 2)
+ (parallel [(const_int 1)
+ (const_int 3)
+ (const_int 5)
+ (const_int 7)]))))))]
+ "TARGET_SSSE3"
+ "pmaddubsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseiadd")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_pmulhrswv8hi3"
+ [(set (match_operand:V8HI 0 "register_operand" "=x")
+ (truncate:V8HI
+ (lshiftrt:V8SI
+ (plus:V8SI
+ (lshiftrt:V8SI
+ (mult:V8SI
+ (sign_extend:V8SI
+ (match_operand:V8HI 1 "nonimmediate_operand" "%0"))
+ (sign_extend:V8SI
+ (match_operand:V8HI 2 "nonimmediate_operand" "xm")))
+ (const_int 14))
+ (const_vector:V8HI [(const_int 1) (const_int 1)
+ (const_int 1) (const_int 1)
+ (const_int 1) (const_int 1)
+ (const_int 1) (const_int 1)]))
+ (const_int 1))))]
+ "TARGET_SSSE3 && ix86_binary_operator_ok (MULT, V8HImode, operands)"
+ "pmulhrsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseimul")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_pmulhrswv4hi3"
+ [(set (match_operand:V4HI 0 "register_operand" "=y")
+ (truncate:V4HI
+ (lshiftrt:V4SI
+ (plus:V4SI
+ (lshiftrt:V4SI
+ (mult:V4SI
+ (sign_extend:V4SI
+ (match_operand:V4HI 1 "nonimmediate_operand" "%0"))
+ (sign_extend:V4SI
+ (match_operand:V4HI 2 "nonimmediate_operand" "ym")))
+ (const_int 14))
+ (const_vector:V4HI [(const_int 1) (const_int 1)
+ (const_int 1) (const_int 1)]))
+ (const_int 1))))]
+ "TARGET_SSSE3 && ix86_binary_operator_ok (MULT, V4HImode, operands)"
+ "pmulhrsw\t{%2, %0|%0, %2}"
+ [(set_attr "type" "sseimul")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_pshufbv16qi3"
+ [(set (match_operand:V16QI 0 "register_operand" "=x")
+ (unspec:V16QI [(match_operand:V16QI 1 "register_operand" "0")
+ (match_operand:V16QI 2 "nonimmediate_operand" "xm")]
+ UNSPEC_PSHUFB))]
+ "TARGET_SSSE3"
+ "pshufb\t{%2, %0|%0, %2}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_pshufbv8qi3"
+ [(set (match_operand:V8QI 0 "register_operand" "=y")
+ (unspec:V8QI [(match_operand:V8QI 1 "register_operand" "0")
+ (match_operand:V8QI 2 "nonimmediate_operand" "ym")]
+ UNSPEC_PSHUFB))]
+ "TARGET_SSSE3"
+ "pshufb\t{%2, %0|%0, %2}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_psign<mode>3"
+ [(set (match_operand:SSEMODE124 0 "register_operand" "=x")
+ (unspec:SSEMODE124 [(match_operand:SSEMODE124 1 "register_operand" "0")
+ (match_operand:SSEMODE124 2 "nonimmediate_operand" "xm")]
+ UNSPEC_PSIGN))]
+ "TARGET_SSSE3"
+ "psign<ssevecsize>\t{%2, %0|%0, %2}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_psign<mode>3"
+ [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+ (unspec:MMXMODEI [(match_operand:MMXMODEI 1 "register_operand" "0")
+ (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")]
+ UNSPEC_PSIGN))]
+ "TARGET_SSSE3"
+ "psign<mmxvecsize>\t{%2, %0|%0, %2}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "DI")])
+
+(define_insn "ssse3_palignrti"
+ [(set (match_operand:TI 0 "register_operand" "=x")
+ (unspec:TI [(match_operand:TI 1 "register_operand" "0")
+ (match_operand:TI 2 "nonimmediate_operand" "xm")
+ (match_operand:SI 3 "const_0_to_255_mul_8_operand" "n")]
+ UNSPEC_PALIGNR))]
+ "TARGET_SSSE3"
+{
+ operands[3] = GEN_INT (INTVAL (operands[3]) / 8);
+ return "palignr\t{%3, %2, %0|%0, %2, %3}";
+}
+ [(set_attr "type" "sseishft")
+ (set_attr "mode" "TI")])
+
+(define_insn "ssse3_palignrdi"
+ [(set (match_operand:DI 0 "register_operand" "=y")
+ (unspec:DI [(match_operand:DI 1 "register_operand" "0")
+ (match_operand:DI 2 "nonimmediate_operand" "ym")
+ (match_operand:SI 3 "const_0_to_255_mul_8_operand" "n")]
+ UNSPEC_PALIGNR))]
+ "TARGET_SSSE3"
+{
+ operands[3] = GEN_INT (INTVAL (operands[3]) / 8);
+ return "palignr\t{%3, %2, %0|%0, %2, %3}";
+}
+ [(set_attr "type" "sseishft")
+ (set_attr "mode" "DI")])
+
+(define_insn "abs<mode>2"
+ [(set (match_operand:SSEMODE124 0 "register_operand" "=x")
+ (abs:SSEMODE124 (match_operand:SSEMODE124 1 "nonimmediate_operand" "xm")))]
+ "TARGET_SSSE3"
+ "pabs<ssevecsize>\t{%1, %0|%0, %1}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "TI")])
+
+(define_insn "abs<mode>2"
+ [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+ (abs:MMXMODEI (match_operand:MMXMODEI 1 "nonimmediate_operand" "ym")))]
+ "TARGET_SSSE3"
+ "pabs<mmxvecsize>\t{%1, %0|%0, %1}";
+ [(set_attr "type" "sselog1")
+ (set_attr "mode" "DI")])
--- /dev/null
+/* Copyright (C) 2006 Free Software Foundation, Inc.
+
+ This file is part of GCC.
+
+ GCC is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ GCC is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with GCC; see the file COPYING. If not, write to
+ the Free Software Foundation, 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+/* As a special exception, if you include this header file into source
+ files compiled by GCC, this header file does not by itself cause
+ the resulting executable to be covered by the GNU General Public
+ License. This exception does not however invalidate any other
+ reasons why the executable file might be covered by the GNU General
+ Public License. */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.1. */
+
+#ifndef _TMMINTRIN_H_INCLUDED
+#define _TMMINTRIN_H_INCLUDED
+
+#ifdef __SSSE3__
+#include <pmmintrin.h>
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadd_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadd_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadds_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadd_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadd_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddd ((__v2si)__X, (__v2si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadds_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsub_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsub_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsubs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsub_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsub_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubd ((__v2si)__X, (__v2si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsubs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_maddubs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pmaddubsw128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_maddubs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pmaddubsw ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_mulhrs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pmulhrsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_mulhrs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pmulhrsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_shuffle_epi8 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pshufb128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_shuffle_pi8 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pshufb ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi8 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignb128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi8 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignb ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignd ((__v2si)__X, (__v2si)__Y);
+}
+
+#define _mm_alignr_epi8(__X, __Y, __N) \
+ ((__m128i)__builtin_ia32_palignr128 ((__v2di) __X, (__v2di) __Y, (__N) * 8))
+
+#define _mm_alignr_pi8(__X, __Y, __N) \
+ ((__m64)__builtin_ia32_palignr ((long long) (__X), (long long) (__Y), (__N) * 8))
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi8 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsb128 ((__v16qi)__X);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi16 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsw128 ((__v8hi)__X);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi32 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsd128 ((__v4si)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi8 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsb ((__v8qi)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi16 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsw ((__v4hi)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi32 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsd ((__v2si)__X);
+}
+
+#endif /* __SSSE3__ */
+
+#endif /* _TMMINTRIN_H_INCLUDED */
+/* Copyright (C) 2006 Free Software Foundation, Inc.
+
+ This file is part of GCC.
+
+ GCC is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ GCC is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with GCC; see the file COPYING. If not, write to
+ the Free Software Foundation, 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+/* As a special exception, if you include this header file into source
+ files compiled by GCC, this header file does not by itself cause
+ the resulting executable to be covered by the GNU General Public
+ License. This exception does not however invalidate any other
+ reasons why the executable file might be covered by the GNU General
+ Public License. */
+
+/* Implemented from the specification included in the Intel C++ Compiler
+ User Guide and Reference, version 9.1. */
+
+#ifndef _TMMINTRIN_H_INCLUDED
+#define _TMMINTRIN_H_INCLUDED
+
+#ifdef __SSSE3__
+#include <pmmintrin.h>
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadd_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadd_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hadds_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phaddsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadd_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadd_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddd ((__v2si)__X, (__v2si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hadds_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phaddsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsub_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsub_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_hsubs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_phsubsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsub_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsub_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubd ((__v2si)__X, (__v2si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_hsubs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_phsubsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_maddubs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pmaddubsw128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_maddubs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pmaddubsw ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_mulhrs_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pmulhrsw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_mulhrs_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pmulhrsw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_shuffle_epi8 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_pshufb128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_shuffle_pi8 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_pshufb ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi8 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignb128 ((__v16qi)__X, (__v16qi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi16 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignw128 ((__v8hi)__X, (__v8hi)__Y);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_sign_epi32 (__m128i __X, __m128i __Y)
+{
+ return (__m128i) __builtin_ia32_psignd128 ((__v4si)__X, (__v4si)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi8 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignb ((__v8qi)__X, (__v8qi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi16 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignw ((__v4hi)__X, (__v4hi)__Y);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_sign_pi32 (__m64 __X, __m64 __Y)
+{
+ return (__m64) __builtin_ia32_psignd ((__v2si)__X, (__v2si)__Y);
+}
+
+#define _mm_alignr_epi8(__X, __Y, __N) \
+ ((__m128i)__builtin_ia32_palignr128 ((__v2di) __X, (__v2di) __Y, (__N) * 8))
+
+#define _mm_alignr_pi8(__X, __Y, __N) \
+ ((__m64)__builtin_ia32_palignr ((long long) (__X), (long long) (__Y), (__N) * 8))
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi8 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsb128 ((__v16qi)__X);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi16 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsw128 ((__v8hi)__X);
+}
+
+static __inline __m128i __attribute__((__always_inline__))
+_mm_abs_epi32 (__m128i __X)
+{
+ return (__m128i) __builtin_ia32_pabsd128 ((__v4si)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi8 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsb ((__v8qi)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi16 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsw ((__v4hi)__X);
+}
+
+static __inline __m64 __attribute__((__always_inline__))
+_mm_abs_pi32 (__m64 __X)
+{
+ return (__m64) __builtin_ia32_pabsd ((__v2si)__X);
+}
+
+#endif /* __SSSE3__ */
+
+#endif /* _TMMINTRIN_H_INCLUDED */
Generates the @code{movddup} machine instruction as a load from memory.
@end table
+The following built-in functions are available when @option{-mssse3} is used.
+All of them generate the machine instruction that is part of the name
+with MMX registers.
+
+@smallexample
+v2si __builtin_ia32_phaddd (v2si, v2si)
+v4hi __builtin_ia32_phaddw (v4hi, v4hi)
+v4hi __builtin_ia32_phaddsw (v4hi, v4hi)
+v2si __builtin_ia32_phsubd (v2si, v2si)
+v4hi __builtin_ia32_phsubw (v4hi, v4hi)
+v4hi __builtin_ia32_phsubsw (v4hi, v4hi)
+v8qi __builtin_ia32_pmaddubsw (v8qi, v8qi)
+v4hi __builtin_ia32_pmulhrsw (v4hi, v4hi)
+v8qi __builtin_ia32_pshufb (v8qi, v8qi)
+v8qi __builtin_ia32_psignb (v8qi, v8qi)
+v2si __builtin_ia32_psignd (v2si, v2si)
+v4hi __builtin_ia32_psignw (v4hi, v4hi)
+long long __builtin_ia32_palignr (long long, long long, int)
+v8qi __builtin_ia32_pabsb (v8qi)
+v2si __builtin_ia32_pabsd (v2si)
+v4hi __builtin_ia32_pabsw (v4hi)
+@end smallexample
+
+The following built-in functions are available when @option{-mssse3} is used.
+All of them generate the machine instruction that is part of the name
+with SSE registers.
+
+@smallexample
+v4si __builtin_ia32_phaddd128 (v4si, v4si)
+v8hi __builtin_ia32_phaddw128 (v8hi, v8hi)
+v8hi __builtin_ia32_phaddsw128 (v8hi, v8hi)
+v4si __builtin_ia32_phsubd128 (v4si, v4si)
+v8hi __builtin_ia32_phsubw128 (v8hi, v8hi)
+v8hi __builtin_ia32_phsubsw128 (v8hi, v8hi)
+v16qi __builtin_ia32_pmaddubsw128 (v16qi, v16qi)
+v8hi __builtin_ia32_pmulhrsw128 (v8hi, v8hi)
+v16qi __builtin_ia32_pshufb128 (v16qi, v16qi)
+v16qi __builtin_ia32_psignb128 (v16qi, v16qi)
+v4si __builtin_ia32_psignd128 (v4si, v4si)
+v8hi __builtin_ia32_psignw128 (v8hi, v8hi)
+v2di __builtin_ia32_palignr (v2di, v2di, int)
+v16qi __builtin_ia32_pabsb128 (v16qi)
+v4si __builtin_ia32_pabsd128 (v4si)
+v8hi __builtin_ia32_pabsw128 (v8hi)
+@end smallexample
+
The following built-in functions are available when @option{-m3dnow} is used.
All of them generate the machine instruction that is part of the name.
-mno-fp-ret-in-387 -msoft-float -msvr3-shlib @gol
-mno-wide-multiply -mrtd -malign-double @gol
-mpreferred-stack-boundary=@var{num} @gol
--mmmx -msse -msse2 -msse3 -m3dnow @gol
+-mmmx -msse -msse2 -msse3 -mssse3 -m3dnow @gol
-mthreads -mno-align-stringops -minline-all-stringops @gol
-mpush-args -maccumulate-outgoing-args -m128bit-long-double @gol
-m96bit-long-double -mregparm=@var{num} -msseregparm @gol
@itemx -mno-sse2
@item -msse3
@itemx -mno-sse3
+@item -mssse3
+@itemx -mno-ssse3
@item -m3dnow
@itemx -mno-3dnow
@opindex mmmx
@opindex m3dnow
@opindex mno-3dnow
These switches enable or disable the use of instructions in the MMX,
-SSE, SSE2 or 3DNow! extended instruction sets. These extensions are
-also available as built-in functions: see @ref{X86 Built-in Functions},
-for details of the functions enabled and disabled by these switches.
+SSE, SSE2, SSE3, SSSE3 or 3DNow! extended instruction sets.
+These extensions are also available as built-in functions: see
+@ref{X86 Built-in Functions}, for details of the functions enabled and
+disabled by these switches.
To have SSE/SSE2 instructions generated automatically from floating-point
code (as opposed to 387 instructions), see @option{-mfpmath=sse}.