procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

sse.md (UNSPEC_MOVU): Remove.

Browse source 
* config/i386/sse.md (UNSPEC_MOVU): Remove.
	(UNSPEC_LOADU): New.
	(UNSPEC_STOREU): Ditto.
	(<sse>_movu<ssemodesuffix><avxsizesuffix>): Split to ...
	(<sse>_loadu<ssemodesuffix><avxsizesuffix>): ... this and ...
	(<sse>_storeu<ssemodesuffix><avxsizesuffix>) ... this.
	(<sse2>_movdqu<avxsizesuffix>): Split to ...
	(<sse2>_loaddqu<avxsizesuffix>): ... this and ...
	(<sse2>_storedqu<avxsizesuffix>): ... this.
	(*sse4_2_pcmpestr_unaligned): Update.
	(*sse4_2_pcmpistr_unaligned): Ditto.

	* config/i386/i386.c (ix86_avx256_split_vector_move_misalign): Use
	gen_avx_load{dqu,ups,upd}256 to load from unaligned memory and
	gen_avx_store{dqu,ups,upd}256 to store to unaligned memory.
	(ix86_expand_vector_move_misalign): Use gen_sse_loadups or
	gen_sse2_load{dqu,upd} to load from unaligned memory and
	gen_sse_loadups or gen_sse2_store{dqu,upd}256 to store to
	unaligned memory.
	(struct builtin_description bdesc_spec) <IX86_BUILTIN_LOADUPS>:
	Use CODE_FOR_sse_loadups.
	<IX86_BUILTIN_LOADUPD>: Use CODE_FOR_sse2_loadupd.
	<IX86_BUILTIN_LOADDQU>: Use CODE_FOR_sse2_loaddqu.
	<IX86_BUILTIN_STOREUPS>: Use CODE_FOR_sse_storeups.
	<IX86_BUILTIN_STOREUPD>: Use CODE_FOR_sse2_storeupd.
	<IX86_BUILTIN_STOREDQU>: Use CODE_FOR_sse2_storedqu.
	<IX86_BUILTIN_LOADUPS256>: Use CODE_FOR_avx_loadups256.
	<IX86_BUILTIN_LOADUPD256>: Use CODE_FOR_avx_loadupd256.
	<IX86_BUILTIN_LOADDQU256>: Use CODE_FOR_avx_loaddqu256.
	<IX86_BUILTIN_STOREUPS256>: Use CODE_FOR_avx_storeups256.
	<IX86_BUILTIN_STOREUPD256>: Use CODE_FOR_avx_storeupd256.
	<IX86_BUILTIN_STOREDQU256>: Use CODE_FOR_avx_storedqu256.

testsuite/ChangeLog:

	* gcc.target/i386/avx256-unaligned-load-1.c: Update asm scan patterns.
	* gcc.target/i386/avx256-unaligned-load-2.c: Ditto.
	* gcc.target/i386/avx256-unaligned-load-3.c: Ditto.
	* gcc.target/i386/avx256-unaligned-load-4.c: Ditto.
	* gcc.target/i386/avx256-unaligned-store-1.c: Ditto.
	* gcc.target/i386/avx256-unaligned-store-2.c: Ditto.
	* gcc.target/i386/avx256-unaligned-store-3.c: Ditto.
	* gcc.target/i386/avx256-unaligned-store-4.c: Ditto.

From-SVN: r192468
This commit is contained in:
Uros Bizjak 2012-10-15 20:42:03 +02:00 • committed by Uros Bizjak
parent 8183317394
commit 860f5e77b7
11 changed files with 183 additions and 65 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

35
gcc/ChangeLog
View file

@ -1,3 +1,38 @@
2012-10-13 Uros Bizjak <ubizjak@gmail.com>
* config/i386/sse.md (UNSPEC_MOVU): Remove.
(UNSPEC_LOADU): New.
(UNSPEC_STOREU): Ditto.
(<sse>_movu<ssemodesuffix><avxsizesuffix>): Split to ...
(<sse>_loadu<ssemodesuffix><avxsizesuffix>): ... this and ...
(<sse>_storeu<ssemodesuffix><avxsizesuffix>) ... this.
(<sse2>_movdqu<avxsizesuffix>): Split to ...
(<sse2>_loaddqu<avxsizesuffix>): ... this and ...
(<sse2>_storedqu<avxsizesuffix>): ... this.
(*sse4_2_pcmpestr_unaligned): Update.
(*sse4_2_pcmpistr_unaligned): Ditto.
* config/i386/i386.c (ix86_avx256_split_vector_move_misalign): Use
gen_avx_load{dqu,ups,upd}256 to load from unaligned memory and
gen_avx_store{dqu,ups,upd}256 to store to unaligned memory.
(ix86_expand_vector_move_misalign): Use gen_sse_loadups or
gen_sse2_load{dqu,upd} to load from unaligned memory and
gen_sse_loadups or gen_sse2_store{dqu,upd}256 to store to
unaligned memory.
(struct builtin_description bdesc_spec) <IX86_BUILTIN_LOADUPS>:
Use CODE_FOR_sse_loadups.
<IX86_BUILTIN_LOADUPD>: Use CODE_FOR_sse2_loadupd.
<IX86_BUILTIN_LOADDQU>: Use CODE_FOR_sse2_loaddqu.
<IX86_BUILTIN_STOREUPS>: Use CODE_FOR_sse_storeups.
<IX86_BUILTIN_STOREUPD>: Use CODE_FOR_sse2_storeupd.
<IX86_BUILTIN_STOREDQU>: Use CODE_FOR_sse2_storedqu.
<IX86_BUILTIN_LOADUPS256>: Use CODE_FOR_avx_loadups256.
<IX86_BUILTIN_LOADUPD256>: Use CODE_FOR_avx_loadupd256.
<IX86_BUILTIN_LOADDQU256>: Use CODE_FOR_avx_loaddqu256.
<IX86_BUILTIN_STOREUPS256>: Use CODE_FOR_avx_storeups256.
<IX86_BUILTIN_STOREUPD256>: Use CODE_FOR_avx_storeupd256.
<IX86_BUILTIN_STOREDQU256>: Use CODE_FOR_avx_storedqu256.
2012-10-15 Dodji Seketeli <dodji@redhat.com>
* alias.c: Cleanup comments.

84
gcc/config/i386/i386.c
View file

@ -16059,7 +16059,8 @@ ix86_avx256_split_vector_move_misalign (rtx op0, rtx op1)
{
rtx m;
rtx (*extract) (rtx, rtx, rtx);
rtx (*move_unaligned) (rtx, rtx);
rtx (*load_unaligned) (rtx, rtx);
rtx (*store_unaligned) (rtx, rtx);
enum machine_mode mode;
switch (GET_MODE (op0))
@ -16068,39 +16069,52 @@ ix86_avx256_split_vector_move_misalign (rtx op0, rtx op1)
gcc_unreachable ();
case V32QImode:
extract = gen_avx_vextractf128v32qi;
move_unaligned = gen_avx_movdqu256;
load_unaligned = gen_avx_loaddqu256;
store_unaligned = gen_avx_storedqu256;
mode = V16QImode;
break;
case V8SFmode:
extract = gen_avx_vextractf128v8sf;
move_unaligned = gen_avx_movups256;
load_unaligned = gen_avx_loadups256;
store_unaligned = gen_avx_storeups256;
mode = V4SFmode;
break;
case V4DFmode:
extract = gen_avx_vextractf128v4df;
move_unaligned = gen_avx_movupd256;
load_unaligned = gen_avx_loadupd256;
store_unaligned = gen_avx_storeupd256;
mode = V2DFmode;
break;
}
if (MEM_P (op1) && TARGET_AVX256_SPLIT_UNALIGNED_LOAD)
if (MEM_P (op1))
{
rtx r = gen_reg_rtx (mode);
m = adjust_address (op1, mode, 0);
emit_move_insn (r, m);
m = adjust_address (op1, mode, 16);
r = gen_rtx_VEC_CONCAT (GET_MODE (op0), r, m);
emit_move_insn (op0, r);
if (TARGET_AVX256_SPLIT_UNALIGNED_LOAD)
{
rtx r = gen_reg_rtx (mode);
m = adjust_address (op1, mode, 0);
emit_move_insn (r, m);
m = adjust_address (op1, mode, 16);
r = gen_rtx_VEC_CONCAT (GET_MODE (op0), r, m);
emit_move_insn (op0, r);
}
else
emit_insn (load_unaligned (op0, op1));
}
else if (MEM_P (op0) && TARGET_AVX256_SPLIT_UNALIGNED_STORE)
else if (MEM_P (op0))
{
m = adjust_address (op0, mode, 0);
emit_insn (extract (m, op1, const0_rtx));
m = adjust_address (op0, mode, 16);
emit_insn (extract (m, op1, const1_rtx));
if (TARGET_AVX256_SPLIT_UNALIGNED_STORE)
{
m = adjust_address (op0, mode, 0);
emit_insn (extract (m, op1, const0_rtx));
m = adjust_address (op0, mode, 16);
emit_insn (extract (m, op1, const1_rtx));
}
else
emit_insn (store_unaligned (op0, op1));
}
else
emit_insn (move_unaligned (op0, op1));
gcc_unreachable ();
}
/* Implement the movmisalign patterns for SSE. Non-SSE modes go
@ -16195,7 +16209,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
op0 = gen_lowpart (V16QImode, op0);
op1 = gen_lowpart (V16QImode, op1);
/* We will eventually emit movups based on insn attributes. */
emit_insn (gen_sse2_movdqu (op0, op1));
emit_insn (gen_sse2_loaddqu (op0, op1));
}
else if (TARGET_SSE2 && mode == V2DFmode)
{
@ -16207,7 +16221,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
|| optimize_function_for_size_p (cfun))
{
/* We will eventually emit movups based on insn attributes. */
emit_insn (gen_sse2_movupd (op0, op1));
emit_insn (gen_sse2_loadupd (op0, op1));
return;
}
@ -16245,7 +16259,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
{
op0 = gen_lowpart (V4SFmode, op0);
op1 = gen_lowpart (V4SFmode, op1);
emit_insn (gen_sse_movups (op0, op1));
emit_insn (gen_sse_loadups (op0, op1));
return;
}
@ -16270,7 +16284,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
op0 = gen_lowpart (V16QImode, op0);
op1 = gen_lowpart (V16QImode, op1);
/* We will eventually emit movups based on insn attributes. */
emit_insn (gen_sse2_movdqu (op0, op1));
emit_insn (gen_sse2_storedqu (op0, op1));
}
else if (TARGET_SSE2 && mode == V2DFmode)
{
@ -16279,7 +16293,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
|| TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL
|| optimize_function_for_size_p (cfun))
/* We will eventually emit movups based on insn attributes. */
emit_insn (gen_sse2_movupd (op0, op1));
emit_insn (gen_sse2_storeupd (op0, op1));
else
{
m = adjust_address (op0, DFmode, 0);
@ -16299,7 +16313,7 @@ ix86_expand_vector_move_misalign (enum machine_mode mode, rtx operands[])
|| optimize_function_for_size_p (cfun))
{
op0 = gen_lowpart (V4SFmode, op0);
emit_insn (gen_sse_movups (op0, op1));
emit_insn (gen_sse_storeups (op0, op1));
}
else
{
@ -26765,9 +26779,9 @@ static const struct builtin_description bdesc_special_args[] =
{ OPTION_MASK_ISA_3DNOW, CODE_FOR_mmx_femms, "__builtin_ia32_femms", IX86_BUILTIN_FEMMS, UNKNOWN, (int) VOID_FTYPE_VOID },
/* SSE */
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_movups, "__builtin_ia32_storeups", IX86_BUILTIN_STOREUPS, UNKNOWN, (int) VOID_FTYPE_PFLOAT_V4SF },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_storeups, "__builtin_ia32_storeups", IX86_BUILTIN_STOREUPS, UNKNOWN, (int) VOID_FTYPE_PFLOAT_V4SF },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_movntv4sf, "__builtin_ia32_movntps", IX86_BUILTIN_MOVNTPS, UNKNOWN, (int) VOID_FTYPE_PFLOAT_V4SF },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_movups, "__builtin_ia32_loadups", IX86_BUILTIN_LOADUPS, UNKNOWN, (int) V4SF_FTYPE_PCFLOAT },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_loadups, "__builtin_ia32_loadups", IX86_BUILTIN_LOADUPS, UNKNOWN, (int) V4SF_FTYPE_PCFLOAT },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_loadhps_exp, "__builtin_ia32_loadhps", IX86_BUILTIN_LOADHPS, UNKNOWN, (int) V4SF_FTYPE_V4SF_PCV2SF },
{ OPTION_MASK_ISA_SSE, CODE_FOR_sse_loadlps_exp, "__builtin_ia32_loadlps", IX86_BUILTIN_LOADLPS, UNKNOWN, (int) V4SF_FTYPE_V4SF_PCV2SF },
@ -26781,14 +26795,14 @@ static const struct builtin_description bdesc_special_args[] =
/* SSE2 */
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_lfence, "__builtin_ia32_lfence", IX86_BUILTIN_LFENCE, UNKNOWN, (int) VOID_FTYPE_VOID },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_mfence, 0, IX86_BUILTIN_MFENCE, UNKNOWN, (int) VOID_FTYPE_VOID },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movupd, "__builtin_ia32_storeupd", IX86_BUILTIN_STOREUPD, UNKNOWN, (int) VOID_FTYPE_PDOUBLE_V2DF },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movdqu, "__builtin_ia32_storedqu", IX86_BUILTIN_STOREDQU, UNKNOWN, (int) VOID_FTYPE_PCHAR_V16QI },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_storeupd, "__builtin_ia32_storeupd", IX86_BUILTIN_STOREUPD, UNKNOWN, (int) VOID_FTYPE_PDOUBLE_V2DF },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_storedqu, "__builtin_ia32_storedqu", IX86_BUILTIN_STOREDQU, UNKNOWN, (int) VOID_FTYPE_PCHAR_V16QI },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movntv2df, "__builtin_ia32_movntpd", IX86_BUILTIN_MOVNTPD, UNKNOWN, (int) VOID_FTYPE_PDOUBLE_V2DF },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movntv2di, "__builtin_ia32_movntdq", IX86_BUILTIN_MOVNTDQ, UNKNOWN, (int) VOID_FTYPE_PV2DI_V2DI },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movntisi, "__builtin_ia32_movnti", IX86_BUILTIN_MOVNTI, UNKNOWN, (int) VOID_FTYPE_PINT_INT },
{ OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_64BIT, CODE_FOR_sse2_movntidi, "__builtin_ia32_movnti64", IX86_BUILTIN_MOVNTI64, UNKNOWN, (int) VOID_FTYPE_PLONGLONG_LONGLONG },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movupd, "__builtin_ia32_loadupd", IX86_BUILTIN_LOADUPD, UNKNOWN, (int) V2DF_FTYPE_PCDOUBLE },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_movdqu, "__builtin_ia32_loaddqu", IX86_BUILTIN_LOADDQU, UNKNOWN, (int) V16QI_FTYPE_PCCHAR },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_loadupd, "__builtin_ia32_loadupd", IX86_BUILTIN_LOADUPD, UNKNOWN, (int) V2DF_FTYPE_PCDOUBLE },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_loaddqu, "__builtin_ia32_loaddqu", IX86_BUILTIN_LOADDQU, UNKNOWN, (int) V16QI_FTYPE_PCCHAR },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_loadhpd_exp, "__builtin_ia32_loadhpd", IX86_BUILTIN_LOADHPD, UNKNOWN, (int) V2DF_FTYPE_V2DF_PCDOUBLE },
{ OPTION_MASK_ISA_SSE2, CODE_FOR_sse2_loadlpd_exp, "__builtin_ia32_loadlpd", IX86_BUILTIN_LOADLPD, UNKNOWN, (int) V2DF_FTYPE_V2DF_PCDOUBLE },
@ -26813,12 +26827,12 @@ static const struct builtin_description bdesc_special_args[] =
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_vbroadcastf128_v4df, "__builtin_ia32_vbroadcastf128_pd256", IX86_BUILTIN_VBROADCASTPD256, UNKNOWN, (int) V4DF_FTYPE_PCV2DF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_vbroadcastf128_v8sf, "__builtin_ia32_vbroadcastf128_ps256", IX86_BUILTIN_VBROADCASTPS256, UNKNOWN, (int) V8SF_FTYPE_PCV4SF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movupd256, "__builtin_ia32_loadupd256", IX86_BUILTIN_LOADUPD256, UNKNOWN, (int) V4DF_FTYPE_PCDOUBLE },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movups256, "__builtin_ia32_loadups256", IX86_BUILTIN_LOADUPS256, UNKNOWN, (int) V8SF_FTYPE_PCFLOAT },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movupd256, "__builtin_ia32_storeupd256", IX86_BUILTIN_STOREUPD256, UNKNOWN, (int) VOID_FTYPE_PDOUBLE_V4DF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movups256, "__builtin_ia32_storeups256", IX86_BUILTIN_STOREUPS256, UNKNOWN, (int) VOID_FTYPE_PFLOAT_V8SF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movdqu256, "__builtin_ia32_loaddqu256", IX86_BUILTIN_LOADDQU256, UNKNOWN, (int) V32QI_FTYPE_PCCHAR },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movdqu256, "__builtin_ia32_storedqu256", IX86_BUILTIN_STOREDQU256, UNKNOWN, (int) VOID_FTYPE_PCHAR_V32QI },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_loadupd256, "__builtin_ia32_loadupd256", IX86_BUILTIN_LOADUPD256, UNKNOWN, (int) V4DF_FTYPE_PCDOUBLE },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_loadups256, "__builtin_ia32_loadups256", IX86_BUILTIN_LOADUPS256, UNKNOWN, (int) V8SF_FTYPE_PCFLOAT },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_storeupd256, "__builtin_ia32_storeupd256", IX86_BUILTIN_STOREUPD256, UNKNOWN, (int) VOID_FTYPE_PDOUBLE_V4DF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_storeups256, "__builtin_ia32_storeups256", IX86_BUILTIN_STOREUPS256, UNKNOWN, (int) VOID_FTYPE_PFLOAT_V8SF },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_loaddqu256, "__builtin_ia32_loaddqu256", IX86_BUILTIN_LOADDQU256, UNKNOWN, (int) V32QI_FTYPE_PCCHAR },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_storedqu256, "__builtin_ia32_storedqu256", IX86_BUILTIN_STOREDQU256, UNKNOWN, (int) VOID_FTYPE_PCHAR_V32QI },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_lddqu256, "__builtin_ia32_lddqu256", IX86_BUILTIN_LDDQU256, UNKNOWN, (int) V32QI_FTYPE_PCCHAR },
{ OPTION_MASK_ISA_AVX, CODE_FOR_avx_movntv4di, "__builtin_ia32_movntdq256", IX86_BUILTIN_MOVNTDQ256, UNKNOWN, (int) VOID_FTYPE_PV4DI_V4DI },

103
gcc/config/i386/sse.md
View file

@ -21,7 +21,8 @@
(define_c_enum "unspec" [
;; SSE
UNSPEC_MOVNT
UNSPEC_MOVU
UNSPEC_LOADU
UNSPEC_STOREU
;; SSE3
UNSPEC_LDDQU
@ -586,12 +587,12 @@
DONE;
})
(define_insn "<sse>_movu<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF 0 "nonimmediate_operand" "=x,m")
(define_insn "<sse>_loadu<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF 0 "register_operand" "=x")
(unspec:VF
[(match_operand:VF 1 "nonimmediate_operand" "xm,x")]
UNSPEC_MOVU))]
"TARGET_SSE && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
[(match_operand:VF 1 "memory_operand" "m")]
UNSPEC_LOADU))]
"TARGET_SSE"
{
switch (get_attr_mode (insn))
{
@ -618,11 +619,79 @@
]
(const_string "<MODE>")))])
(define_insn "<sse2>_movdqu<avxsizesuffix>"
[(set (match_operand:VI1 0 "nonimmediate_operand" "=x,m")
(unspec:VI1 [(match_operand:VI1 1 "nonimmediate_operand" "xm,x")]
UNSPEC_MOVU))]
"TARGET_SSE2 && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
(define_insn "<sse>_storeu<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF 0 "memory_operand" "=m")
(unspec:VF
[(match_operand:VF 1 "register_operand" "x")]
UNSPEC_STOREU))]
"TARGET_SSE"
{
switch (get_attr_mode (insn))
{
case MODE_V8SF:
case MODE_V4SF:
return "%vmovups\t{%1, %0|%0, %1}";
default:
return "%vmovu<ssemodesuffix>\t{%1, %0|%0, %1}";
}
}
[(set_attr "type" "ssemov")
(set_attr "movu" "1")
(set_attr "prefix" "maybe_vex")
(set (attr "mode")
(cond [(match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL")
(const_string "<ssePSmode>")
(and (eq_attr "alternative" "1")
(match_test "TARGET_SSE_TYPELESS_STORES"))
(const_string "<ssePSmode>")
(match_test "TARGET_AVX")
(const_string "<MODE>")
(match_test "optimize_function_for_size_p (cfun)")
(const_string "V4SF")
]
(const_string "<MODE>")))])
(define_insn "<sse2>_loaddqu<avxsizesuffix>"
[(set (match_operand:VI1 0 "register_operand" "=x")
(unspec:VI1 [(match_operand:VI1 1 "memory_operand" "m")]
UNSPEC_LOADU))]
"TARGET_SSE2"
{
switch (get_attr_mode (insn))
{
case MODE_V8SF:
case MODE_V4SF:
return "%vmovups\t{%1, %0|%0, %1}";
default:
return "%vmovdqu\t{%1, %0|%0, %1}";
}
}
[(set_attr "type" "ssemov")
(set_attr "movu" "1")
(set (attr "prefix_data16")
(if_then_else
(match_test "TARGET_AVX")
(const_string "*")
(const_string "1")))
(set_attr "prefix" "maybe_vex")
(set (attr "mode")
(cond [(match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL")
(const_string "<ssePSmode>")
(and (eq_attr "alternative" "1")
(match_test "TARGET_SSE_TYPELESS_STORES"))
(const_string "<ssePSmode>")
(match_test "TARGET_AVX")
(const_string "<sseinsnmode>")
(match_test "optimize_function_for_size_p (cfun)")
(const_string "V4SF")
]
(const_string "<sseinsnmode>")))])
(define_insn "<sse2>_storedqu<avxsizesuffix>"
[(set (match_operand:VI1 0 "memory_operand" "=m")
(unspec:VI1 [(match_operand:VI1 1 "register_operand" "x")]
UNSPEC_STOREU))]
"TARGET_SSE2"
{
switch (get_attr_mode (insn))
{
@ -9307,7 +9376,7 @@
(match_operand:SI 3 "register_operand" "a")
(unspec:V16QI
[(match_operand:V16QI 4 "memory_operand" "m")]
UNSPEC_MOVU)
UNSPEC_LOADU)
(match_operand:SI 5 "register_operand" "d")
(match_operand:SI 6 "const_0_to_255_operand" "n")]
UNSPEC_PCMPESTR))
@ -9315,7 +9384,7 @@
(unspec:V16QI
[(match_dup 2)
(match_dup 3)
(unspec:V16QI [(match_dup 4)] UNSPEC_MOVU)
(unspec:V16QI [(match_dup 4)] UNSPEC_LOADU)
(match_dup 5)
(match_dup 6)]
UNSPEC_PCMPESTR))
@ -9323,7 +9392,7 @@
(unspec:CC
[(match_dup 2)
(match_dup 3)
(unspec:V16QI [(match_dup 4)] UNSPEC_MOVU)
(unspec:V16QI [(match_dup 4)] UNSPEC_LOADU)
(match_dup 5)
(match_dup 6)]
UNSPEC_PCMPESTR))]
@ -9498,19 +9567,19 @@
[(match_operand:V16QI 2 "register_operand" "x")
(unspec:V16QI
[(match_operand:V16QI 3 "memory_operand" "m")]
UNSPEC_MOVU)
UNSPEC_LOADU)
(match_operand:SI 4 "const_0_to_255_operand" "n")]
UNSPEC_PCMPISTR))
(set (match_operand:V16QI 1 "register_operand" "=Yz")
(unspec:V16QI
[(match_dup 2)
(unspec:V16QI [(match_dup 3)] UNSPEC_MOVU)
(unspec:V16QI [(match_dup 3)] UNSPEC_LOADU)
(match_dup 4)]
UNSPEC_PCMPISTR))
(set (reg:CC FLAGS_REG)
(unspec:CC
[(match_dup 2)
(unspec:V16QI [(match_dup 3)] UNSPEC_MOVU)
(unspec:V16QI [(match_dup 3)] UNSPEC_LOADU)
(match_dup 4)]
UNSPEC_PCMPISTR))]
"TARGET_SSE4_2

4
gcc/testsuite/gcc.target/i386/avx256-unaligned-load-1.c
View file

@ -14,6 +14,6 @@ avx_test (void)
c[i] = a[i] * b[i+3];
}
/* { dg-final { scan-assembler-not "avx_movups256/1" } } */
/* { dg-final { scan-assembler "sse_movups/1" } } */
/* { dg-final { scan-assembler-not "avx_loadups256" } } */
/* { dg-final { scan-assembler "sse_loadups" } } */
/* { dg-final { scan-assembler "vinsertf128" } } */

4
gcc/testsuite/gcc.target/i386/avx256-unaligned-load-2.c
View file

@ -24,6 +24,6 @@ avx_test (void)
}
}
/* { dg-final { scan-assembler-not "avx_movdqu256/1" } } */
/* { dg-final { scan-assembler "sse2_movdqu/1" } } */
/* { dg-final { scan-assembler-not "avx_loaddqu256" } } */
/* { dg-final { scan-assembler "sse2_loaddqu" } } */
/* { dg-final { scan-assembler "vinsert.128" } } */

4
gcc/testsuite/gcc.target/i386/avx256-unaligned-load-3.c
View file

@ -14,6 +14,6 @@ avx_test (void)
c[i] = a[i] * b[i+3];
}
/* { dg-final { scan-assembler-not "avx_movupd256/1" } } */
/* { dg-final { scan-assembler "sse2_movupd/1" } } */
/* { dg-final { scan-assembler-not "avx_loadupd256" } } */
/* { dg-final { scan-assembler "sse2_loadupd" } } */
/* { dg-final { scan-assembler "vinsertf128" } } */

4
gcc/testsuite/gcc.target/i386/avx256-unaligned-load-4.c
View file

@ -14,6 +14,6 @@ avx_test (void)
b[i] = a[i+3] * 2;
}
/* { dg-final { scan-assembler "avx_movups256/1" } } */
/* { dg-final { scan-assembler-not "avx_movups/1" } } */
/* { dg-final { scan-assembler "avx_loadups256" } } */
/* { dg-final { scan-assembler-not "sse_loadups" } } */
/* { dg-final { scan-assembler-not "vinsertf128" } } */

2
gcc/testsuite/gcc.target/i386/avx256-unaligned-store-1.c
View file

@ -17,6 +17,6 @@ avx_test (void)
d[i] = c[i] * 20.0;
}
/* { dg-final { scan-assembler-not "avx_movups256/2" } } */
/* { dg-final { scan-assembler-not "avx_storeups256" } } */
/* { dg-final { scan-assembler "vmovups.*\\*movv4sf_internal/3" } } */
/* { dg-final { scan-assembler "vextractf128" } } */

2
gcc/testsuite/gcc.target/i386/avx256-unaligned-store-2.c
View file

@ -24,6 +24,6 @@ avx_test (void)
}
}
/* { dg-final { scan-assembler-not "avx_movdqu256/2" } } */
/* { dg-final { scan-assembler-not "avx_storedqu256" } } */
/* { dg-final { scan-assembler "vmovdqu.*\\*movv16qi_internal/3" } } */
/* { dg-final { scan-assembler "vextract.128" } } */

2
gcc/testsuite/gcc.target/i386/avx256-unaligned-store-3.c
View file

@ -17,6 +17,6 @@ avx_test (void)
d[i] = c[i] * 20.0;
}
/* { dg-final { scan-assembler-not "avx_movupd256/2" } } */
/* { dg-final { scan-assembler-not "avx_storeupd256" } } */
/* { dg-final { scan-assembler "vmovupd.*\\*movv2df_internal/3" } } */
/* { dg-final { scan-assembler "vextractf128" } } */

4
gcc/testsuite/gcc.target/i386/avx256-unaligned-store-4.c
View file

@ -14,7 +14,7 @@ avx_test (void)
b[i+3] = a[i] * c[i];
}
/* { dg-final { scan-assembler "avx_movups256/2" } } */
/* { dg-final { scan-assembler-not "avx_movups/2" } } */
/* { dg-final { scan-assembler "avx_storeups256" } } */
/* { dg-final { scan-assembler-not "sse_storeups" } } */
/* { dg-final { scan-assembler-not "\\*avx_movv4sf_internal/3" } } */
/* { dg-final { scan-assembler-not "vextractf128" } } */