This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] PR target/68991: Add vector_memory_operand and "Bm" constraint
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: Uros Bizjak <ubizjak at gmail dot com>, Richard Biener <richard dot guenther at gmail dot com>, Kirill Yukhin <kirill dot yukhin at gmail dot com>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, Vladimir Makarov <vmakarov at redhat dot com>
- Date: Tue, 5 Jan 2016 10:58:50 -0800
- Subject: Re: [PATCH] PR target/68991: Add vector_memory_operand and "Bm" constraint
- Authentication-results: sourceware.org; auth=none
- References: <20151230205333 dot GA1877 at gmail dot com> <CAFULd4bAa2+Jn10-wnLzTVkUWVh1KOgpCA8LM1DMUDAyNb2baA at mail dot gmail dot com> <CAMe9rOrYmTPPFE3mre7kcWEGoHJx32ogQ60J1gPryo2d593nKg at mail dot gmail dot com> <CAFULd4Y5HmdwT_N9n_ZDcj6W54tuT37Z-D+k_VYM2qentfVDyw at mail dot gmail dot com> <9193C1A5-1430-49CF-9F47-CB673218BA30 at gmail dot com> <CAMe9rOoKtKo_xV0Nn7ufNowXjfkS4XsLnifHPVTCUCCobFM8ew at mail dot gmail dot com> <CAMe9rOreR+8cxrTvw-f-WtMUVq6F_XhDJ6yYFwxjLd5orWiExw at mail dot gmail dot com> <CAFULd4YKifWmi0hMuTZALd1xzEKQAZ_G=q4vRJBQU_iEvYm5kA at mail dot gmail dot com> <CAMe9rOqJY-uOd5sQBqphv9NakPTE5oog0FwoQf3-oyRPeZzOHg at mail dot gmail dot com> <CAMe9rOpOhuBk46DHFkdJymGcD89yDH+GGeongQSkK4txN_e-Uw at mail dot gmail dot com> <20160105081120 dot GQ18720 at tucnak dot redhat dot com> <CAMe9rOq5FT9HafS6Eq9_4qSoEtZVKrwmiX=0SDPPxBf7wPvUcw at mail dot gmail dot com>
On Tue, Jan 5, 2016 at 4:32 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Tue, Jan 5, 2016 at 12:11 AM, Jakub Jelinek <jakub@redhat.com> wrote:
>> On Mon, Jan 04, 2016 at 03:25:48PM -0800, H.J. Lu wrote:
>>> LRA is fine. I should use
>>>
>>> (define_memory_constraint "Bm"
>>> "@internal Vector memory operand."
>>> (match_operand 0 "vector_memory_operand"))
>>>
>>> instead of
>>>
>>> (define_constraint "Bm"
>>> "@internal Vector memory operand."
>>> (match_operand 0 "vector_memory_operand"))
>>
>> I don't think so. At least the documentation says that
>> define_memory_constraint is for MEM constraints where if they are not
>> satisfied they can be made to satisfy by forcing the address into a
>> register. But that is not the case here, if a MEM is misaligned, no
>> equivalent changes to the XEXP (mem, 0) will make it aligned.
>>
>
> You are right and *mov<mode>_internal must use the 'm' constraint
> so that LRA won't keep generating the same reload for
>
> (insn 353 322 323 8 (set (reg:V4SF 192)
> (reg:V4SF 201 [192])) 1226 {*movv4sf_internal}
> (nil))
>
> until
>
> x.i: In function \u2018foo\u2019:
> x.i:29:1: internal compiler error: Max. number of generated reload
> insns per insn is achieved (90)
>
> }
> ^
>
> 0xc0d635 lra_constraints(bool)
> /export/gnu/import/git/sources/gcc/gcc/lra-constraints.c:4336
> 0xbf9854 lra(_IO_FILE*)
> /export/gnu/import/git/sources/gcc/gcc/lra.c:2277
> 0xba6489 do_reload
> /export/gnu/import/git/sources/gcc/gcc/ira.c:5385
> 0xba683c execute
> /export/gnu/import/git/sources/gcc/gcc/ira.c:5556
>
>
Here are the updated patches. I didn't change SSE
*mov<mode>_internal. Tested on x86-64. OK for
trunk?
--
H.J.
From 74f90ab1f38a7baf7503974130a699a1bd622b87 Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Sat, 19 Dec 2015 06:51:44 -0800
Subject: [PATCH 1/2] Add vector_memory_operand and "Bm" constraint
SSE vector arithmetic and logic instructions only accept aligned memory
operand. This patch adds vector_memory_operand and "Bm" constraint for
aligned SSE memory operand. They are applied to SSE plusminus and
any_logic patterns.
gcc/
PR target/68991
* config/i386/constraints.md (Bm): New constraint.
* config/i386/predicates.md (vector_memory_operand): New
predicate.
* config/i386/sse.md: Replace xm with xBm in plusminus and
any_logic patterns.
gcc/testsuite/
PR target/68991
* g++.dg/pr68991-1.C: New test.
* g++.dg/pr68991-2.C: Likewise.
---
gcc/config/i386/constraints.md | 5 +
gcc/config/i386/predicates.md | 7 ++
gcc/config/i386/sse.md | 18 ++--
gcc/testsuite/g++.dg/pr68991-1.C | 191 +++++++++++++++++++++++++++++++++++++++
gcc/testsuite/g++.dg/pr68991-2.C | 191 +++++++++++++++++++++++++++++++++++++++
5 files changed, 403 insertions(+), 9 deletions(-)
create mode 100644 gcc/testsuite/g++.dg/pr68991-1.C
create mode 100644 gcc/testsuite/g++.dg/pr68991-2.C
diff --git a/gcc/config/i386/constraints.md b/gcc/config/i386/constraints.md
index b3db133..bac9d66 100644
--- a/gcc/config/i386/constraints.md
+++ b/gcc/config/i386/constraints.md
@@ -148,6 +148,7 @@
;; We use the B prefix to denote any number of internal operands:
;; f FLAGS_REG
;; g GOT memory operand.
+;; m Vector memory operand
;; s Sibcall memory operand, not valid for TARGET_X32
;; w Call memory operand, not valid for TARGET_X32
;; z Constant call address operand.
@@ -160,6 +161,10 @@
"@internal GOT memory operand."
(match_operand 0 "GOT_memory_operand"))
+(define_constraint "Bm"
+ "@internal Vector memory operand."
+ (match_operand 0 "vector_memory_operand"))
+
(define_constraint "Bs"
"@internal Sibcall memory operand."
(ior (and (not (match_test "TARGET_X32"))
diff --git a/gcc/config/i386/predicates.md b/gcc/config/i386/predicates.md
index 6f0159b..33b3594 100644
--- a/gcc/config/i386/predicates.md
+++ b/gcc/config/i386/predicates.md
@@ -951,6 +951,13 @@
(match_test "INTEGRAL_MODE_P (GET_MODE (op))")
(match_test "op == CONSTM1_RTX (GET_MODE (op))")))
+; Return true when OP is operand acceptable for vector memory operand.
+; Only AVX can have misaligned memory operand.
+(define_predicate "vector_memory_operand"
+ (and (match_operand 0 "memory_operand")
+ (ior (match_test "TARGET_AVX")
+ (match_test "MEM_ALIGN (op) >= GET_MODE_ALIGNMENT (mode)"))))
+
; Return true when OP is operand acceptable for standard SSE move.
(define_predicate "vector_move_operand"
(ior (match_operand 0 "nonimmediate_operand")
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index c3cd199..ca3a831 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -1676,7 +1676,7 @@
[(set (match_operand:VF 0 "register_operand" "=x,v")
(plusminus:VF
(match_operand:VF 1 "<round_nimm_predicate>" "<comm>0,v")
- (match_operand:VF 2 "<round_nimm_predicate>" "xm,<round_constraint>")))]
+ (match_operand:VF 2 "<round_nimm_predicate>" "xBm,<round_constraint>")))]
"TARGET_SSE && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands) && <mask_mode512bit_condition> && <round_mode512bit_condition>"
"@
<plusminus_mnemonic><ssemodesuffix>\t{%2, %0|%0, %2}
@@ -1691,7 +1691,7 @@
(vec_merge:VF_128
(plusminus:VF_128
(match_operand:VF_128 1 "register_operand" "0,v")
- (match_operand:VF_128 2 "nonimmediate_operand" "xm,<round_constraint>"))
+ (match_operand:VF_128 2 "nonimmediate_operand" "xBm,<round_constraint>"))
(match_dup 1)
(const_int 1)))]
"TARGET_SSE"
@@ -2424,7 +2424,7 @@
(vec_concat:V2SF
(plusminus:SF
(vec_select:SF
- (match_operand:V4SF 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V4SF 2 "nonimmediate_operand" "xBm,xm")
(parallel [(const_int 0)]))
(vec_select:SF (match_dup 2) (parallel [(const_int 1)])))
(plusminus:SF
@@ -3240,7 +3240,7 @@
[(set (match_operand:VF_128_256 0 "register_operand" "=x,v")
(any_logic:VF_128_256
(match_operand:VF_128_256 1 "nonimmediate_operand" "%0,v")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VF_128_256 2 "nonimmediate_operand" "xBm,vm")))]
"TARGET_SSE && <mask_avx512vl_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
@@ -3489,7 +3489,7 @@
[(set (match_operand:TF 0 "register_operand" "=x,x")
(any_logic:TF
(match_operand:TF 1 "nonimmediate_operand" "%0,x")
- (match_operand:TF 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:TF 2 "nonimmediate_operand" "xBm,xm")))]
"TARGET_SSE
&& ix86_binary_operator_ok (<CODE>, TFmode, operands)"
{
@@ -9653,7 +9653,7 @@
[(set (match_operand:VI_AVX2 0 "register_operand" "=x,v")
(plusminus:VI_AVX2
(match_operand:VI_AVX2 1 "nonimmediate_operand" "<comm>0,v")
- (match_operand:VI_AVX2 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VI_AVX2 2 "nonimmediate_operand" "xBm,vm")))]
"TARGET_SSE2
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
"@
@@ -9706,7 +9706,7 @@
[(set (match_operand:VI12_AVX2 0 "register_operand" "=x,v")
(sat_plusminus:VI12_AVX2
(match_operand:VI12_AVX2 1 "nonimmediate_operand" "<comm>0,v")
- (match_operand:VI12_AVX2 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VI12_AVX2 2 "nonimmediate_operand" "xBm,vm")))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
"@
@@ -11361,7 +11361,7 @@
[(set (match_operand:VI48_AVX_AVX512F 0 "register_operand" "=x,v")
(any_logic:VI48_AVX_AVX512F
(match_operand:VI48_AVX_AVX512F 1 "nonimmediate_operand" "%0,v")
- (match_operand:VI48_AVX_AVX512F 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VI48_AVX_AVX512F 2 "nonimmediate_operand" "xBm,vm")))]
"TARGET_SSE && <mask_mode512bit_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
@@ -11457,7 +11457,7 @@
[(set (match_operand:VI12_AVX_AVX512F 0 "register_operand" "=x,v")
(any_logic: VI12_AVX_AVX512F
(match_operand:VI12_AVX_AVX512F 1 "nonimmediate_operand" "%0,v")
- (match_operand:VI12_AVX_AVX512F 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VI12_AVX_AVX512F 2 "nonimmediate_operand" "xBm,vm")))]
"TARGET_SSE && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
static char buf[64];
diff --git a/gcc/testsuite/g++.dg/pr68991-1.C b/gcc/testsuite/g++.dg/pr68991-1.C
new file mode 100644
index 0000000..744d13c
--- /dev/null
+++ b/gcc/testsuite/g++.dg/pr68991-1.C
@@ -0,0 +1,191 @@
+// { dg-do compile { target i?86-*-* x86_64-*-* } }
+// { dg-options "-std=c++11 -O3 -msse2 -mno-avx -fno-exceptions -fno-rtti -fdump-rtl-final" }
+
+typedef unsigned int size_type;
+
+#define _GLIBCXX_BITSET_BITS_PER_WORD (__CHAR_BIT__ * __SIZEOF_INT__)
+#define _GLIBCXX_BITSET_WORDS(__n) \
+ ((__n) / _GLIBCXX_BITSET_BITS_PER_WORD + \
+ ((__n) % _GLIBCXX_BITSET_BITS_PER_WORD == 0 ? 0 : 1))
+
+namespace std
+{
+ template<size_type _Nw>
+ struct _Base_bitset
+ {
+ typedef unsigned int _WordT;
+ _WordT _M_w[_Nw];
+
+ _WordT&
+ _M_hiword()
+ { return _M_w[_Nw - 1]; }
+
+ void
+ _M_do_and(const _Base_bitset<_Nw>& __x)
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ _M_w[__i] &= __x._M_w[__i];
+ }
+
+ void
+ _M_do_flip()
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ _M_w[__i] = ~_M_w[__i];
+ }
+
+ bool
+ _M_is_equal(const _Base_bitset<_Nw>& __x) const
+ {
+ for (size_type __i = 0; __i < _Nw; ++__i)
+ if (_M_w[__i] != __x._M_w[__i])
+ return false;
+ return true;
+ }
+
+ bool
+ _M_is_any() const
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ if (_M_w[__i] != static_cast<_WordT>(0))
+ return true;
+ return false;
+ }
+ };
+
+ template<size_type _Extrabits>
+ struct _Sanitize
+ {
+ typedef unsigned int _WordT;
+
+ static void
+ _S_do_sanitize(_WordT& __val)
+ { __val &= ~((~static_cast<_WordT>(0)) << _Extrabits); }
+ };
+
+ template<size_type _Nb>
+ class bitset
+ : private _Base_bitset<_GLIBCXX_BITSET_WORDS(_Nb)>
+ {
+ private:
+ typedef _Base_bitset<_GLIBCXX_BITSET_WORDS(_Nb)> _Base;
+ typedef unsigned int _WordT;
+
+ void
+ _M_do_sanitize()
+ {
+ typedef _Sanitize<_Nb % _GLIBCXX_BITSET_BITS_PER_WORD> __sanitize_type;
+ __sanitize_type::_S_do_sanitize(this->_M_hiword());
+ }
+
+ public:
+ class reference
+ {
+ friend class bitset;
+
+ _WordT* _M_wp;
+ size_type _M_bpos;
+
+ public:
+ reference&
+ flip()
+ {
+ *_M_wp ^= _Base::_S_maskbit(_M_bpos);
+ return *this;
+ }
+ };
+
+ bitset<_Nb>&
+ operator&=(const bitset<_Nb>& __rhs)
+ {
+ this->_M_do_and(__rhs);
+ return *this;
+ }
+
+ bitset<_Nb>&
+ flip()
+ {
+ this->_M_do_flip();
+ this->_M_do_sanitize();
+ return *this;
+ }
+
+ bitset<_Nb>
+ operator~() const
+ { return bitset<_Nb>(*this).flip(); }
+
+ bool
+ operator==(const bitset<_Nb>& __rhs) const
+ { return this->_M_is_equal(__rhs); }
+
+ bool
+ any() const
+ { return this->_M_is_any(); }
+ };
+
+ template<size_type _Nb>
+ inline bitset<_Nb>
+ operator&(const bitset<_Nb>& __x, const bitset<_Nb>& __y)
+ {
+ bitset<_Nb> __result(__x);
+ __result &= __y;
+ return __result;
+ }
+}
+template<typename T>
+class ArrayRef {
+public:
+ typedef const T *iterator;
+
+private:
+ const T *Data;
+ size_type Length;
+
+public:
+ iterator begin() const { return Data; }
+ iterator end() const { return Data + Length; }
+};
+
+const unsigned MAX_SUBTARGET_FEATURES = 128;
+class FeatureBitset : public std::bitset<MAX_SUBTARGET_FEATURES> {
+};
+
+struct SubtargetFeatureKV {
+ FeatureBitset Value;
+ FeatureBitset Implies;
+};
+
+struct SubtargetInfoKV {
+ const void *Value;
+};
+class SubtargetFeatures {
+public:
+ FeatureBitset ToggleFeature(FeatureBitset Bits,
+ const SubtargetFeatureKV *,
+ ArrayRef<SubtargetFeatureKV> FeatureTable);
+};
+
+static
+void ClearImpliedBits(FeatureBitset &Bits,
+ const SubtargetFeatureKV *FeatureEntry,
+ ArrayRef<SubtargetFeatureKV> FeatureTable) {
+ for (auto &FE : FeatureTable) {
+ if ((FE.Implies & FeatureEntry->Value).any()) {
+ Bits &= ~FE.Value;
+ ClearImpliedBits(Bits, &FE, FeatureTable);
+ }
+ }
+}
+
+FeatureBitset
+SubtargetFeatures::ToggleFeature(FeatureBitset Bits,
+ const SubtargetFeatureKV *FeatureEntry,
+ ArrayRef<SubtargetFeatureKV> FeatureTable) {
+ if ((Bits & FeatureEntry->Value) == FeatureEntry->Value) {
+ Bits &= ~FeatureEntry->Value;
+ ClearImpliedBits(Bits, FeatureEntry, FeatureTable);
+ }
+ return Bits;
+}
+
+// { dg-final { scan-rtl-dump-not "S16 A32\[^\n\]*\\\*xorv4si3" "final" } }
diff --git a/gcc/testsuite/g++.dg/pr68991-2.C b/gcc/testsuite/g++.dg/pr68991-2.C
new file mode 100644
index 0000000..a3c59f0
--- /dev/null
+++ b/gcc/testsuite/g++.dg/pr68991-2.C
@@ -0,0 +1,191 @@
+// { dg-do compile { target i?86-*-* x86_64-*-* } }
+// { dg-options "-std=c++11 -O3 -msse2 -mno-avx -fno-exceptions -fno-rtti -fdump-rtl-final" }
+
+typedef unsigned int size_type;
+
+#define _GLIBCXX_BITSET_BITS_PER_WORD (__CHAR_BIT__ * __SIZEOF_INT__)
+#define _GLIBCXX_BITSET_WORDS(__n) \
+ ((__n) / _GLIBCXX_BITSET_BITS_PER_WORD + \
+ ((__n) % _GLIBCXX_BITSET_BITS_PER_WORD == 0 ? 0 : 1))
+
+namespace std
+{
+ template<size_type _Nw>
+ struct _Base_bitset
+ {
+ typedef unsigned int _WordT;
+ _WordT _M_w[_Nw];
+
+ _WordT&
+ _M_hiword()
+ { return _M_w[_Nw - 1]; }
+
+ void
+ _M_do_and(const _Base_bitset<_Nw>& __x)
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ _M_w[__i] += __x._M_w[__i];
+ }
+
+ void
+ _M_do_flip()
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ _M_w[__i] = ~_M_w[__i];
+ }
+
+ bool
+ _M_is_equal(const _Base_bitset<_Nw>& __x) const
+ {
+ for (size_type __i = 0; __i < _Nw; ++__i)
+ if (_M_w[__i] != __x._M_w[__i])
+ return false;
+ return true;
+ }
+
+ bool
+ _M_is_any() const
+ {
+ for (size_type __i = 0; __i < _Nw; __i++)
+ if (_M_w[__i] != static_cast<_WordT>(0))
+ return true;
+ return false;
+ }
+ };
+
+ template<size_type _Extrabits>
+ struct _Sanitize
+ {
+ typedef unsigned int _WordT;
+
+ static void
+ _S_do_sanitize(_WordT& __val)
+ { __val &= ~((~static_cast<_WordT>(0)) << _Extrabits); }
+ };
+
+ template<size_type _Nb>
+ class bitset
+ : private _Base_bitset<_GLIBCXX_BITSET_WORDS(_Nb)>
+ {
+ private:
+ typedef _Base_bitset<_GLIBCXX_BITSET_WORDS(_Nb)> _Base;
+ typedef unsigned int _WordT;
+
+ void
+ _M_do_sanitize()
+ {
+ typedef _Sanitize<_Nb % _GLIBCXX_BITSET_BITS_PER_WORD> __sanitize_type;
+ __sanitize_type::_S_do_sanitize(this->_M_hiword());
+ }
+
+ public:
+ class reference
+ {
+ friend class bitset;
+
+ _WordT* _M_wp;
+ size_type _M_bpos;
+
+ public:
+ reference&
+ flip()
+ {
+ *_M_wp ^= _Base::_S_maskbit(_M_bpos);
+ return *this;
+ }
+ };
+
+ bitset<_Nb>&
+ operator&=(const bitset<_Nb>& __rhs)
+ {
+ this->_M_do_and(__rhs);
+ return *this;
+ }
+
+ bitset<_Nb>&
+ flip()
+ {
+ this->_M_do_flip();
+ this->_M_do_sanitize();
+ return *this;
+ }
+
+ bitset<_Nb>
+ operator~() const
+ { return bitset<_Nb>(*this).flip(); }
+
+ bool
+ operator==(const bitset<_Nb>& __rhs) const
+ { return this->_M_is_equal(__rhs); }
+
+ bool
+ any() const
+ { return this->_M_is_any(); }
+ };
+
+ template<size_type _Nb>
+ inline bitset<_Nb>
+ operator&(const bitset<_Nb>& __x, const bitset<_Nb>& __y)
+ {
+ bitset<_Nb> __result(__x);
+ __result &= __y;
+ return __result;
+ }
+}
+template<typename T>
+class ArrayRef {
+public:
+ typedef const T *iterator;
+
+private:
+ const T *Data;
+ size_type Length;
+
+public:
+ iterator begin() const { return Data; }
+ iterator end() const { return Data + Length; }
+};
+
+const unsigned MAX_SUBTARGET_FEATURES = 128;
+class FeatureBitset : public std::bitset<MAX_SUBTARGET_FEATURES> {
+};
+
+struct SubtargetFeatureKV {
+ FeatureBitset Value;
+ FeatureBitset Implies;
+};
+
+struct SubtargetInfoKV {
+ const void *Value;
+};
+class SubtargetFeatures {
+public:
+ FeatureBitset ToggleFeature(FeatureBitset Bits,
+ const SubtargetFeatureKV *,
+ ArrayRef<SubtargetFeatureKV> FeatureTable);
+};
+
+static
+void ClearImpliedBits(FeatureBitset &Bits,
+ const SubtargetFeatureKV *FeatureEntry,
+ ArrayRef<SubtargetFeatureKV> FeatureTable) {
+ for (auto &FE : FeatureTable) {
+ if ((FE.Implies & FeatureEntry->Value).any()) {
+ Bits &= ~FE.Value;
+ ClearImpliedBits(Bits, &FE, FeatureTable);
+ }
+ }
+}
+
+FeatureBitset
+SubtargetFeatures::ToggleFeature(FeatureBitset Bits,
+ const SubtargetFeatureKV *FeatureEntry,
+ ArrayRef<SubtargetFeatureKV> FeatureTable) {
+ if ((Bits & FeatureEntry->Value) == FeatureEntry->Value) {
+ Bits &= ~FeatureEntry->Value;
+ ClearImpliedBits(Bits, FeatureEntry, FeatureTable);
+ }
+ return Bits;
+}
+
+// { dg-final { scan-rtl-dump-not "S16 A32\[^\n\]*\\\*addv4si3" "final" } }
--
2.5.0
From 6c402dad0af335eb2660090537e7bde9328798ed Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Sat, 2 Jan 2016 07:47:30 -0800
Subject: [PATCH 2/2] Use vector_operand on SSE with 16b memory operand
Add vector_operand, which is vector_memory_operand or register_operand,
and use it, instead of nonimmediate_operand, in SSE patterns with 16-byte
memory operand.
gcc/
PR target/68991
* config/i386/i386.c (ix86_expand_vector_logical_operator):
Replace nonimmediate_operand with vector_operand.
* config/i386/predicates.md (vector_operand): New predicate.
(general_vector_operand): Replace nonimmediate_operand with
vector_operand.
* config/i386/sse.md: Replace nonimmediate_operand with
vector_operand and m constraint with Bm constraint on SSE
patterns with 16-byte memory operand.
* config/i386/subst.md (round_nimm_predicate): Replace
nonimmediate_operand with vector_operand.
(round_saeonly_nimm_predicate): Likewise.
(round_saeonly_nimm_scalar_predicate): New.
gcc/testsuite/
PR target/68991
* gcc.target/i386/pr68991.c: New test.
---
gcc/config/i386/i386.c | 8 +-
gcc/config/i386/predicates.md | 9 +-
gcc/config/i386/sse.md | 475 ++++++++++++++++----------------
gcc/config/i386/subst.md | 5 +-
gcc/testsuite/gcc.target/i386/pr68991.c | 32 +++
5 files changed, 284 insertions(+), 245 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/i386/pr68991.c
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index db29b0a..c6c66c7 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -19359,11 +19359,11 @@ ix86_expand_vector_logical_operator (enum rtx_code code, machine_mode mode,
{
op1 = operands[1];
op2 = SUBREG_REG (operands[2]);
- if (!nonimmediate_operand (op2, GET_MODE (dst)))
+ if (!vector_operand (op2, GET_MODE (dst)))
op2 = force_reg (GET_MODE (dst), op2);
}
op1 = SUBREG_REG (op1);
- if (!nonimmediate_operand (op1, GET_MODE (dst)))
+ if (!vector_operand (op1, GET_MODE (dst)))
op1 = force_reg (GET_MODE (dst), op1);
emit_insn (gen_rtx_SET (dst,
gen_rtx_fmt_ee (code, GET_MODE (dst),
@@ -19374,9 +19374,9 @@ ix86_expand_vector_logical_operator (enum rtx_code code, machine_mode mode,
break;
}
}
- if (!nonimmediate_operand (operands[1], mode))
+ if (!vector_operand (operands[1], mode))
operands[1] = force_reg (mode, operands[1]);
- if (!nonimmediate_operand (operands[2], mode))
+ if (!vector_operand (operands[2], mode))
operands[2] = force_reg (mode, operands[2]);
ix86_fixup_binary_operands_no_copy (code, mode, operands);
emit_insn (gen_rtx_SET (operands[0],
diff --git a/gcc/config/i386/predicates.md b/gcc/config/i386/predicates.md
index 33b3594..14e80d9 100644
--- a/gcc/config/i386/predicates.md
+++ b/gcc/config/i386/predicates.md
@@ -958,6 +958,11 @@
(ior (match_test "TARGET_AVX")
(match_test "MEM_ALIGN (op) >= GET_MODE_ALIGNMENT (mode)"))))
+; Return true when OP is register_operand or vector_memory_operand.
+(define_predicate "vector_operand"
+ (ior (match_operand 0 "register_operand")
+ (match_operand 0 "vector_memory_operand")))
+
; Return true when OP is operand acceptable for standard SSE move.
(define_predicate "vector_move_operand"
(ior (match_operand 0 "nonimmediate_operand")
@@ -1598,9 +1603,9 @@
return val == ((low << 8) | low);
})
-;; Return true if OP is nonimmediate_operand or CONST_VECTOR.
+;; Return true if OP is vector_operand or CONST_VECTOR.
(define_predicate "general_vector_operand"
- (ior (match_operand 0 "nonimmediate_operand")
+ (ior (match_operand 0 "vector_operand")
(match_code "const_vector")))
;; Return true if OP is either -1 constant or stored in register.
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index ca3a831..c21cc0e 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -1629,8 +1629,8 @@
(define_insn_and_split "*absneg<mode>2"
[(set (match_operand:VF 0 "register_operand" "=x,x,v,v")
(match_operator:VF 3 "absneg_operator"
- [(match_operand:VF 1 "nonimmediate_operand" "0, xm, v, m")]))
- (use (match_operand:VF 2 "nonimmediate_operand" "xm, 0, vm,v"))]
+ [(match_operand:VF 1 "vector_operand" "0, xBm,v, m")]))
+ (use (match_operand:VF 2 "vector_operand" "xBm,0, vm,v"))]
"TARGET_SSE"
"#"
"&& reload_completed"
@@ -1691,7 +1691,7 @@
(vec_merge:VF_128
(plusminus:VF_128
(match_operand:VF_128 1 "register_operand" "0,v")
- (match_operand:VF_128 2 "nonimmediate_operand" "xBm,<round_constraint>"))
+ (match_operand:VF_128 2 "vector_operand" "xBm,<round_constraint>"))
(match_dup 1)
(const_int 1)))]
"TARGET_SSE"
@@ -1715,7 +1715,7 @@
[(set (match_operand:VF 0 "register_operand" "=x,v")
(mult:VF
(match_operand:VF 1 "<round_nimm_predicate>" "%0,v")
- (match_operand:VF 2 "<round_nimm_predicate>" "xm,<round_constraint>")))]
+ (match_operand:VF 2 "<round_nimm_predicate>" "xBm,<round_constraint>")))]
"TARGET_SSE && ix86_binary_operator_ok (MULT, <MODE>mode, operands) && <mask_mode512bit_condition> && <round_mode512bit_condition>"
"@
mul<ssemodesuffix>\t{%2, %0|%0, %2}
@@ -1731,7 +1731,7 @@
(vec_merge:VF_128
(multdiv:VF_128
(match_operand:VF_128 1 "register_operand" "0,v")
- (match_operand:VF_128 2 "nonimmediate_operand" "xm,<round_constraint>"))
+ (match_operand:VF_128 2 "vector_operand" "xBm,<round_constraint>"))
(match_dup 1)
(const_int 1)))]
"TARGET_SSE"
@@ -1747,14 +1747,14 @@
(define_expand "div<mode>3"
[(set (match_operand:VF2 0 "register_operand")
(div:VF2 (match_operand:VF2 1 "register_operand")
- (match_operand:VF2 2 "nonimmediate_operand")))]
+ (match_operand:VF2 2 "vector_operand")))]
"TARGET_SSE2"
"ix86_fixup_binary_operands_no_copy (DIV, <MODE>mode, operands);")
(define_expand "div<mode>3"
[(set (match_operand:VF1 0 "register_operand")
(div:VF1 (match_operand:VF1 1 "register_operand")
- (match_operand:VF1 2 "nonimmediate_operand")))]
+ (match_operand:VF1 2 "vector_operand")))]
"TARGET_SSE"
{
ix86_fixup_binary_operands_no_copy (DIV, <MODE>mode, operands);
@@ -1774,7 +1774,7 @@
[(set (match_operand:VF 0 "register_operand" "=x,v")
(div:VF
(match_operand:VF 1 "register_operand" "0,v")
- (match_operand:VF 2 "<round_nimm_predicate>" "xm,<round_constraint>")))]
+ (match_operand:VF 2 "<round_nimm_predicate>" "xBm,<round_constraint>")))]
"TARGET_SSE && <mask_mode512bit_condition> && <round_mode512bit_condition>"
"@
div<ssemodesuffix>\t{%2, %0|%0, %2}
@@ -1787,7 +1787,7 @@
(define_insn "<sse>_rcp<mode>2"
[(set (match_operand:VF1_128_256 0 "register_operand" "=x")
(unspec:VF1_128_256
- [(match_operand:VF1_128_256 1 "nonimmediate_operand" "xm")] UNSPEC_RCP))]
+ [(match_operand:VF1_128_256 1 "vector_operand" "xBm")] UNSPEC_RCP))]
"TARGET_SSE"
"%vrcpps\t{%1, %0|%0, %1}"
[(set_attr "type" "sse")
@@ -1842,12 +1842,12 @@
(define_expand "sqrt<mode>2"
[(set (match_operand:VF2 0 "register_operand")
- (sqrt:VF2 (match_operand:VF2 1 "nonimmediate_operand")))]
+ (sqrt:VF2 (match_operand:VF2 1 "vector_operand")))]
"TARGET_SSE2")
(define_expand "sqrt<mode>2"
[(set (match_operand:VF1 0 "register_operand")
- (sqrt:VF1 (match_operand:VF1 1 "nonimmediate_operand")))]
+ (sqrt:VF1 (match_operand:VF1 1 "vector_operand")))]
"TARGET_SSE"
{
if (TARGET_SSE_MATH
@@ -1876,7 +1876,7 @@
[(set (match_operand:VF_128 0 "register_operand" "=x,v")
(vec_merge:VF_128
(sqrt:VF_128
- (match_operand:VF_128 1 "nonimmediate_operand" "xm,<round_constraint>"))
+ (match_operand:VF_128 1 "vector_operand" "xBm,<round_constraint>"))
(match_operand:VF_128 2 "register_operand" "0,v")
(const_int 1)))]
"TARGET_SSE"
@@ -1893,7 +1893,7 @@
(define_expand "rsqrt<mode>2"
[(set (match_operand:VF1_128_256 0 "register_operand")
(unspec:VF1_128_256
- [(match_operand:VF1_128_256 1 "nonimmediate_operand")] UNSPEC_RSQRT))]
+ [(match_operand:VF1_128_256 1 "vector_operand")] UNSPEC_RSQRT))]
"TARGET_SSE_MATH"
{
ix86_emit_swsqrtsf (operands[0], operands[1], <MODE>mode, true);
@@ -1903,7 +1903,7 @@
(define_insn "<sse>_rsqrt<mode>2"
[(set (match_operand:VF1_128_256 0 "register_operand" "=x")
(unspec:VF1_128_256
- [(match_operand:VF1_128_256 1 "nonimmediate_operand" "xm")] UNSPEC_RSQRT))]
+ [(match_operand:VF1_128_256 1 "vector_operand" "xBm")] UNSPEC_RSQRT))]
"TARGET_SSE"
"%vrsqrtps\t{%1, %0|%0, %1}"
[(set_attr "type" "sse")
@@ -1972,7 +1972,7 @@
[(set (match_operand:VF 0 "register_operand" "=x,v")
(smaxmin:VF
(match_operand:VF 1 "<round_saeonly_nimm_predicate>" "%0,v")
- (match_operand:VF 2 "<round_saeonly_nimm_predicate>" "xm,<round_saeonly_constraint>")))]
+ (match_operand:VF 2 "<round_saeonly_nimm_predicate>" "xBm,<round_saeonly_constraint>")))]
"TARGET_SSE && flag_finite_math_only
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)
&& <mask_mode512bit_condition> && <round_saeonly_mode512bit_condition>"
@@ -1989,7 +1989,7 @@
[(set (match_operand:VF 0 "register_operand" "=x,v")
(smaxmin:VF
(match_operand:VF 1 "register_operand" "0,v")
- (match_operand:VF 2 "<round_saeonly_nimm_predicate>" "xm,<round_saeonly_constraint>")))]
+ (match_operand:VF 2 "<round_saeonly_nimm_predicate>" "xBm,<round_saeonly_constraint>")))]
"TARGET_SSE && !flag_finite_math_only
&& <mask_mode512bit_condition> && <round_saeonly_mode512bit_condition>"
"@
@@ -2006,7 +2006,7 @@
(vec_merge:VF_128
(smaxmin:VF_128
(match_operand:VF_128 1 "register_operand" "0,v")
- (match_operand:VF_128 2 "nonimmediate_operand" "xm,<round_saeonly_constraint>"))
+ (match_operand:VF_128 2 "vector_operand" "xBm,<round_saeonly_constraint>"))
(match_dup 1)
(const_int 1)))]
"TARGET_SSE"
@@ -2026,10 +2026,10 @@
;; presence of -0.0 and NaN.
(define_insn "*ieee_smin<mode>3"
- [(set (match_operand:VF 0 "register_operand" "=v,v")
+ [(set (match_operand:VF 0 "register_operand" "=x,v")
(unspec:VF
[(match_operand:VF 1 "register_operand" "0,v")
- (match_operand:VF 2 "nonimmediate_operand" "vm,vm")]
+ (match_operand:VF 2 "vector_operand" "xBm,vm")]
UNSPEC_IEEE_MIN))]
"TARGET_SSE"
"@
@@ -2041,10 +2041,10 @@
(set_attr "mode" "<MODE>")])
(define_insn "*ieee_smax<mode>3"
- [(set (match_operand:VF 0 "register_operand" "=v,v")
+ [(set (match_operand:VF 0 "register_operand" "=x,v")
(unspec:VF
[(match_operand:VF 1 "register_operand" "0,v")
- (match_operand:VF 2 "nonimmediate_operand" "vm,vm")]
+ (match_operand:VF 2 "vector_operand" "xBm,vm")]
UNSPEC_IEEE_MAX))]
"TARGET_SSE"
"@
@@ -2074,7 +2074,7 @@
(vec_merge:V2DF
(minus:V2DF
(match_operand:V2DF 1 "register_operand" "0,x")
- (match_operand:V2DF 2 "nonimmediate_operand" "xm,xm"))
+ (match_operand:V2DF 2 "vector_operand" "xBm,xm"))
(plus:V2DF (match_dup 1) (match_dup 2))
(const_int 1)))]
"TARGET_SSE3"
@@ -2106,7 +2106,7 @@
(vec_merge:V4SF
(minus:V4SF
(match_operand:V4SF 1 "register_operand" "0,x")
- (match_operand:V4SF 2 "nonimmediate_operand" "xm,xm"))
+ (match_operand:V4SF 2 "vector_operand" "xBm,xm"))
(plus:V4SF (match_dup 1) (match_dup 2))
(const_int 5)))]
"TARGET_SSE3"
@@ -2124,10 +2124,10 @@
(match_operator:VF_128_256 6 "addsub_vm_operator"
[(minus:VF_128_256
(match_operand:VF_128_256 1 "register_operand")
- (match_operand:VF_128_256 2 "nonimmediate_operand"))
+ (match_operand:VF_128_256 2 "vector_operand"))
(plus:VF_128_256
- (match_operand:VF_128_256 3 "nonimmediate_operand")
- (match_operand:VF_128_256 4 "nonimmediate_operand"))
+ (match_operand:VF_128_256 3 "vector_operand")
+ (match_operand:VF_128_256 4 "vector_operand"))
(match_operand 5 "const_int_operand")]))]
"TARGET_SSE3
&& can_create_pseudo_p ()
@@ -2145,11 +2145,11 @@
[(set (match_operand:VF_128_256 0 "register_operand")
(match_operator:VF_128_256 6 "addsub_vm_operator"
[(plus:VF_128_256
- (match_operand:VF_128_256 1 "nonimmediate_operand")
- (match_operand:VF_128_256 2 "nonimmediate_operand"))
+ (match_operand:VF_128_256 1 "vector_operand")
+ (match_operand:VF_128_256 2 "vector_operand"))
(minus:VF_128_256
(match_operand:VF_128_256 3 "register_operand")
- (match_operand:VF_128_256 4 "nonimmediate_operand"))
+ (match_operand:VF_128_256 4 "vector_operand"))
(match_operand 5 "const_int_operand")]))]
"TARGET_SSE3
&& can_create_pseudo_p ()
@@ -2175,10 +2175,10 @@
[(vec_concat:<ssedoublemode>
(minus:VF_128_256
(match_operand:VF_128_256 1 "register_operand")
- (match_operand:VF_128_256 2 "nonimmediate_operand"))
+ (match_operand:VF_128_256 2 "vector_operand"))
(plus:VF_128_256
- (match_operand:VF_128_256 3 "nonimmediate_operand")
- (match_operand:VF_128_256 4 "nonimmediate_operand")))
+ (match_operand:VF_128_256 3 "vector_operand")
+ (match_operand:VF_128_256 4 "vector_operand")))
(match_parallel 5 "addsub_vs_parallel"
[(match_operand 6 "const_int_operand")])]))]
"TARGET_SSE3
@@ -2208,11 +2208,11 @@
(match_operator:VF_128_256 7 "addsub_vs_operator"
[(vec_concat:<ssedoublemode>
(plus:VF_128_256
- (match_operand:VF_128_256 1 "nonimmediate_operand")
- (match_operand:VF_128_256 2 "nonimmediate_operand"))
+ (match_operand:VF_128_256 1 "vector_operand")
+ (match_operand:VF_128_256 2 "vector_operand"))
(minus:VF_128_256
(match_operand:VF_128_256 3 "register_operand")
- (match_operand:VF_128_256 4 "nonimmediate_operand")))
+ (match_operand:VF_128_256 4 "vector_operand")))
(match_parallel 5 "addsub_vs_parallel"
[(match_operand 6 "const_int_operand")])]))]
"TARGET_SSE3
@@ -2274,7 +2274,7 @@
(vec_select:DF (match_dup 1) (parallel [(const_int 1)])))
(plus:DF
(vec_select:DF
- (match_operand:V2DF 2 "nonimmediate_operand")
+ (match_operand:V2DF 2 "vector_operand")
(parallel [(const_int 0)]))
(vec_select:DF (match_dup 2) (parallel [(const_int 1)])))))]
"TARGET_SSE3")
@@ -2291,7 +2291,7 @@
(parallel [(match_operand:SI 4 "const_0_to_1_operand")])))
(plus:DF
(vec_select:DF
- (match_operand:V2DF 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V2DF 2 "vector_operand" "xBm,xm")
(parallel [(match_operand:SI 5 "const_0_to_1_operand")]))
(vec_select:DF
(match_dup 2)
@@ -2317,7 +2317,7 @@
(vec_select:DF (match_dup 1) (parallel [(const_int 1)])))
(minus:DF
(vec_select:DF
- (match_operand:V2DF 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V2DF 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0)]))
(vec_select:DF (match_dup 2) (parallel [(const_int 1)])))))]
"TARGET_SSE3"
@@ -2424,7 +2424,7 @@
(vec_concat:V2SF
(plusminus:SF
(vec_select:SF
- (match_operand:V4SF 2 "nonimmediate_operand" "xBm,xm")
+ (match_operand:V4SF 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0)]))
(vec_select:SF (match_dup 2) (parallel [(const_int 1)])))
(plusminus:SF
@@ -2650,7 +2650,7 @@
[(set (match_operand:VF_128_256 0 "register_operand" "=x,x")
(match_operator:VF_128_256 3 "sse_comparison_operator"
[(match_operand:VF_128_256 1 "register_operand" "%0,x")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "xm,xm")]))]
+ (match_operand:VF_128_256 2 "vector_operand" "xBm,xm")]))]
"TARGET_SSE
&& GET_RTX_CLASS (GET_CODE (operands[3])) == RTX_COMM_COMPARE"
"@
@@ -2666,7 +2666,7 @@
[(set (match_operand:VF_128_256 0 "register_operand" "=x,x")
(match_operator:VF_128_256 3 "sse_comparison_operator"
[(match_operand:VF_128_256 1 "register_operand" "0,x")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "xm,xm")]))]
+ (match_operand:VF_128_256 2 "vector_operand" "xBm,xm")]))]
"TARGET_SSE"
"@
cmp%D3<ssemodesuffix>\t{%2, %0|%0, %2}
@@ -2682,7 +2682,7 @@
(vec_merge:VF_128
(match_operator:VF_128 3 "sse_comparison_operator"
[(match_operand:VF_128 1 "register_operand" "0,x")
- (match_operand:VF_128 2 "nonimmediate_operand" "xm,xm")])
+ (match_operand:VF_128 2 "vector_operand" "xBm,xm")])
(match_dup 1)
(const_int 1)))]
"TARGET_SSE"
@@ -2815,7 +2815,7 @@
(match_operand:<ssevecmode> 0 "register_operand" "v")
(parallel [(const_int 0)]))
(vec_select:MODEF
- (match_operand:<ssevecmode> 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:<ssevecmode> 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"SSE_FLOAT_MODE_P (<MODE>mode)"
"%vcomi<ssemodesuffix>\t{<round_saeonly_op2>%1, %0|%0, %<iptr>1<round_saeonly_op2>}"
@@ -2835,7 +2835,7 @@
(match_operand:<ssevecmode> 0 "register_operand" "v")
(parallel [(const_int 0)]))
(vec_select:MODEF
- (match_operand:<ssevecmode> 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:<ssevecmode> 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"SSE_FLOAT_MODE_P (<MODE>mode)"
"%vucomi<ssemodesuffix>\t{<round_saeonly_op2>%1, %0|%0, %<iptr>1<round_saeonly_op2>}"
@@ -2888,7 +2888,7 @@
[(set (match_operand:<sseintvecmode> 0 "register_operand")
(match_operator:<sseintvecmode> 1 ""
[(match_operand:VI124_128 2 "register_operand")
- (match_operand:VI124_128 3 "nonimmediate_operand")]))]
+ (match_operand:VI124_128 3 "vector_operand")]))]
"TARGET_SSE2"
{
bool ok = ix86_expand_int_vec_cmp (operands);
@@ -2900,7 +2900,7 @@
[(set (match_operand:V2DI 0 "register_operand")
(match_operator:V2DI 1 ""
[(match_operand:V2DI 2 "register_operand")
- (match_operand:V2DI 3 "nonimmediate_operand")]))]
+ (match_operand:V2DI 3 "vector_operand")]))]
"TARGET_SSE4_2"
{
bool ok = ix86_expand_int_vec_cmp (operands);
@@ -2924,7 +2924,7 @@
[(set (match_operand:<sseintvecmode> 0 "register_operand")
(match_operator:<sseintvecmode> 1 ""
[(match_operand:VF_128 2 "register_operand")
- (match_operand:VF_128 3 "nonimmediate_operand")]))]
+ (match_operand:VF_128 3 "vector_operand")]))]
"TARGET_SSE"
{
bool ok = ix86_expand_fp_vec_cmp (operands);
@@ -2972,7 +2972,7 @@
[(set (match_operand:<sseintvecmode> 0 "register_operand")
(match_operator:<sseintvecmode> 1 ""
[(match_operand:VI124_128 2 "register_operand")
- (match_operand:VI124_128 3 "nonimmediate_operand")]))]
+ (match_operand:VI124_128 3 "vector_operand")]))]
"TARGET_SSE2"
{
bool ok = ix86_expand_int_vec_cmp (operands);
@@ -2984,7 +2984,7 @@
[(set (match_operand:V2DI 0 "register_operand")
(match_operator:V2DI 1 ""
[(match_operand:V2DI 2 "register_operand")
- (match_operand:V2DI 3 "nonimmediate_operand")]))]
+ (match_operand:V2DI 3 "vector_operand")]))]
"TARGET_SSE4_2"
{
bool ok = ix86_expand_int_vec_cmp (operands);
@@ -3030,8 +3030,8 @@
[(set (match_operand:V_128 0 "register_operand")
(if_then_else:V_128
(match_operator 3 ""
- [(match_operand:VF_128 4 "nonimmediate_operand")
- (match_operand:VF_128 5 "nonimmediate_operand")])
+ [(match_operand:VF_128 4 "vector_operand")
+ (match_operand:VF_128 5 "vector_operand")])
(match_operand:V_128 1 "general_operand")
(match_operand:V_128 2 "general_operand")))]
"TARGET_SSE
@@ -3075,7 +3075,7 @@
(define_expand "vcond_mask_<mode><sseintvecmodelower>"
[(set (match_operand:VI124_128 0 "register_operand")
(vec_merge:VI124_128
- (match_operand:VI124_128 1 "nonimmediate_operand")
+ (match_operand:VI124_128 1 "vector_operand")
(match_operand:VI124_128 2 "vector_move_operand")
(match_operand:<sseintvecmode> 3 "register_operand")))]
"TARGET_SSE2"
@@ -3088,7 +3088,7 @@
(define_expand "vcond_mask_v2div2di"
[(set (match_operand:V2DI 0 "register_operand")
(vec_merge:V2DI
- (match_operand:V2DI 1 "nonimmediate_operand")
+ (match_operand:V2DI 1 "vector_operand")
(match_operand:V2DI 2 "vector_move_operand")
(match_operand:V2DI 3 "register_operand")))]
"TARGET_SSE4_2"
@@ -3114,7 +3114,7 @@
(define_expand "vcond_mask_<mode><sseintvecmodelower>"
[(set (match_operand:VF_128 0 "register_operand")
(vec_merge:VF_128
- (match_operand:VF_128 1 "nonimmediate_operand")
+ (match_operand:VF_128 1 "vector_operand")
(match_operand:VF_128 2 "vector_move_operand")
(match_operand:<sseintvecmode> 3 "register_operand")))]
"TARGET_SSE"
@@ -3135,7 +3135,7 @@
(and:VF_128_256
(not:VF_128_256
(match_operand:VF_128_256 1 "register_operand" "0,v"))
- (match_operand:VF_128_256 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VF_128_256 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE && <mask_avx512vl_condition>"
{
static char buf[128];
@@ -3223,8 +3223,8 @@
(define_expand "<code><mode>3<mask_name>"
[(set (match_operand:VF_128_256 0 "register_operand")
(any_logic:VF_128_256
- (match_operand:VF_128_256 1 "nonimmediate_operand")
- (match_operand:VF_128_256 2 "nonimmediate_operand")))]
+ (match_operand:VF_128_256 1 "vector_operand")
+ (match_operand:VF_128_256 2 "vector_operand")))]
"TARGET_SSE && <mask_avx512vl_condition>"
"ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
@@ -3239,8 +3239,8 @@
(define_insn "*<code><mode>3<mask_name>"
[(set (match_operand:VF_128_256 0 "register_operand" "=x,v")
(any_logic:VF_128_256
- (match_operand:VF_128_256 1 "nonimmediate_operand" "%0,v")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "xBm,vm")))]
+ (match_operand:VF_128_256 1 "vector_operand" "%0,v")
+ (match_operand:VF_128_256 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE && <mask_avx512vl_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
@@ -3328,10 +3328,10 @@
[(set (match_dup 4)
(and:VF
(not:VF (match_dup 3))
- (match_operand:VF 1 "nonimmediate_operand")))
+ (match_operand:VF 1 "vector_operand")))
(set (match_dup 5)
(and:VF (match_dup 3)
- (match_operand:VF 2 "nonimmediate_operand")))
+ (match_operand:VF 2 "vector_operand")))
(set (match_operand:VF 0 "register_operand")
(ior:VF (match_dup 4) (match_dup 5)))]
"TARGET_SSE"
@@ -3393,7 +3393,7 @@
[(set (match_operand:TF 0 "register_operand" "=x,x")
(and:TF
(not:TF (match_operand:TF 1 "register_operand" "0,x"))
- (match_operand:TF 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:TF 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE"
{
static char buf[32];
@@ -3480,16 +3480,16 @@
(define_expand "<code>tf3"
[(set (match_operand:TF 0 "register_operand")
(any_logic:TF
- (match_operand:TF 1 "nonimmediate_operand")
- (match_operand:TF 2 "nonimmediate_operand")))]
+ (match_operand:TF 1 "vector_operand")
+ (match_operand:TF 2 "vector_operand")))]
"TARGET_SSE"
"ix86_fixup_binary_operands_no_copy (<CODE>, TFmode, operands);")
(define_insn "*<code>tf3"
[(set (match_operand:TF 0 "register_operand" "=x,x")
(any_logic:TF
- (match_operand:TF 1 "nonimmediate_operand" "%0,x")
- (match_operand:TF 2 "nonimmediate_operand" "xBm,xm")))]
+ (match_operand:TF 1 "vector_operand" "%0,x")
+ (match_operand:TF 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE
&& ix86_binary_operator_ok (<CODE>, TFmode, operands)"
{
@@ -4374,7 +4374,7 @@
[(set (match_operand:SI 0 "register_operand" "=r,r")
(fix:SI
(vec_select:SF
- (match_operand:V4SF 1 "<round_saeonly_nimm_predicate>" "v,<round_saeonly_constraint2>")
+ (match_operand:V4SF 1 "<round_saeonly_nimm_scalar_predicate>" "v,<round_saeonly_constraint2>")
(parallel [(const_int 0)]))))]
"TARGET_SSE"
"%vcvttss2si\t{<round_saeonly_op2>%1, %0|%0, %k1<round_saeonly_op2>}"
@@ -4390,7 +4390,7 @@
[(set (match_operand:DI 0 "register_operand" "=r,r")
(fix:DI
(vec_select:SF
- (match_operand:V4SF 1 "<round_saeonly_nimm_predicate>" "v,<round_saeonly_constraint>")
+ (match_operand:V4SF 1 "<round_saeonly_nimm_scalar_predicate>" "v,<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"TARGET_SSE && TARGET_64BIT"
"%vcvttss2si{q}\t{<round_saeonly_op2>%1, %0|%0, %k1<round_saeonly_op2>}"
@@ -4479,7 +4479,7 @@
(define_insn "<sse2_avx_avx512f>_fix_notrunc<sf2simodelower><mode><mask_name>"
[(set (match_operand:VI4_AVX 0 "register_operand" "=v")
(unspec:VI4_AVX
- [(match_operand:<ssePSmode> 1 "nonimmediate_operand" "vm")]
+ [(match_operand:<ssePSmode> 1 "vector_operand" "vBm")]
UNSPEC_FIX_NOTRUNC))]
"TARGET_SSE2 && <mask_mode512bit_condition>"
"%vcvtps2dq\t{%1, %0<mask_operand2>|%0<mask_operand2>, %1}"
@@ -4581,7 +4581,7 @@
(define_insn "fix_truncv4sfv4si2<mask_name>"
[(set (match_operand:V4SI 0 "register_operand" "=v")
- (fix:V4SI (match_operand:V4SF 1 "nonimmediate_operand" "vm")))]
+ (fix:V4SI (match_operand:V4SF 1 "vector_operand" "vBm")))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
"%vcvttps2dq\t{%1, %0<mask_operand2>|%0<mask_operand2>, %1}"
[(set_attr "type" "ssecvt")
@@ -4732,7 +4732,7 @@
[(set (match_operand:SI 0 "register_operand" "=r")
(unsigned_fix:SI
(vec_select:SF
- (match_operand:V4SF 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:V4SF 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"TARGET_AVX512F"
"vcvttss2usi\t{<round_saeonly_op2>%1, %0|%0, %1<round_saeonly_op2>}"
@@ -4744,7 +4744,7 @@
[(set (match_operand:DI 0 "register_operand" "=r")
(unsigned_fix:DI
(vec_select:SF
- (match_operand:V4SF 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:V4SF 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"TARGET_AVX512F && TARGET_64BIT"
"vcvttss2usi\t{<round_saeonly_op2>%1, %0|%0, %1<round_saeonly_op2>}"
@@ -4782,7 +4782,7 @@
[(set (match_operand:SI 0 "register_operand" "=r")
(unsigned_fix:SI
(vec_select:DF
- (match_operand:V2DF 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:V2DF 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"TARGET_AVX512F"
"vcvttsd2usi\t{<round_saeonly_op2>%1, %0|%0, %1<round_saeonly_op2>}"
@@ -4794,7 +4794,7 @@
[(set (match_operand:DI 0 "register_operand" "=r")
(unsigned_fix:DI
(vec_select:DF
- (match_operand:V2DF 1 "<round_saeonly_nimm_predicate>" "<round_saeonly_constraint>")
+ (match_operand:V2DF 1 "<round_saeonly_nimm_scalar_predicate>" "<round_saeonly_constraint>")
(parallel [(const_int 0)]))))]
"TARGET_AVX512F && TARGET_64BIT"
"vcvttsd2usi\t{<round_saeonly_op2>%1, %0|%0, %1<round_saeonly_op2>}"
@@ -4867,7 +4867,7 @@
[(set (match_operand:SI 0 "register_operand" "=r,r")
(fix:SI
(vec_select:DF
- (match_operand:V2DF 1 "<round_saeonly_nimm_predicate>" "v,<round_saeonly_constraint2>")
+ (match_operand:V2DF 1 "<round_saeonly_nimm_scalar_predicate>" "v,<round_saeonly_constraint2>")
(parallel [(const_int 0)]))))]
"TARGET_SSE2"
"%vcvttsd2si\t{<round_saeonly_op2>%1, %0|%0, %q1<round_saeonly_op2>}"
@@ -4884,7 +4884,7 @@
[(set (match_operand:DI 0 "register_operand" "=r,r")
(fix:DI
(vec_select:DF
- (match_operand:V2DF 1 "<round_saeonly_nimm_predicate>" "v,<round_saeonly_constraint2>")
+ (match_operand:V2DF 1 "<round_saeonly_nimm_scalar_predicate>" "v,<round_saeonly_constraint2>")
(parallel [(const_int 0)]))))]
"TARGET_SSE2 && TARGET_64BIT"
"%vcvttsd2si{q}\t{<round_saeonly_op2>%1, %0|%0, %q1<round_saeonly_op2>}"
@@ -5081,7 +5081,7 @@
(define_insn "sse2_cvtpd2dq<mask_name>"
[(set (match_operand:V4SI 0 "register_operand" "=v")
(vec_concat:V4SI
- (unspec:V2SI [(match_operand:V2DF 1 "nonimmediate_operand" "vm")]
+ (unspec:V2SI [(match_operand:V2DF 1 "vector_operand" "vBm")]
UNSPEC_FIX_NOTRUNC)
(const_vector:V2SI [(const_int 0) (const_int 0)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
@@ -5242,7 +5242,7 @@
(define_insn "sse2_cvttpd2dq<mask_name>"
[(set (match_operand:V4SI 0 "register_operand" "=v")
(vec_concat:V4SI
- (fix:V2SI (match_operand:V2DF 1 "nonimmediate_operand" "vm"))
+ (fix:V2SI (match_operand:V2DF 1 "vector_operand" "vBm"))
(const_vector:V2SI [(const_int 0) (const_int 0)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
{
@@ -5285,7 +5285,7 @@
(vec_merge:V2DF
(float_extend:V2DF
(vec_select:V2SF
- (match_operand:V4SF 2 "<round_saeonly_nimm_predicate>" "x,m,<round_saeonly_constraint>")
+ (match_operand:V4SF 2 "<round_saeonly_nimm_scalar_predicate>" "x,m,<round_saeonly_constraint>")
(parallel [(const_int 0) (const_int 1)])))
(match_operand:V2DF 1 "register_operand" "0,0,v")
(const_int 1)))]
@@ -5328,7 +5328,7 @@
[(set (match_operand:V4SF 0 "register_operand")
(vec_concat:V4SF
(float_truncate:V2SF
- (match_operand:V2DF 1 "nonimmediate_operand"))
+ (match_operand:V2DF 1 "vector_operand"))
(match_dup 2)))]
"TARGET_SSE2"
"operands[2] = CONST0_RTX (V2SFmode);")
@@ -5338,7 +5338,7 @@
(vec_merge:V4SF
(vec_concat:V4SF
(float_truncate:V2SF
- (match_operand:V2DF 1 "nonimmediate_operand"))
+ (match_operand:V2DF 1 "vector_operand"))
(match_dup 4))
(match_operand:V4SF 2 "register_operand")
(match_operand:QI 3 "register_operand")))]
@@ -5349,7 +5349,7 @@
[(set (match_operand:V4SF 0 "register_operand" "=v")
(vec_concat:V4SF
(float_truncate:V2SF
- (match_operand:V2DF 1 "nonimmediate_operand" "vm"))
+ (match_operand:V2DF 1 "vector_operand" "vBm"))
(match_operand:V2SF 2 "const0_operand")))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
{
@@ -5478,7 +5478,7 @@
[(set (match_operand:V2DF 0 "register_operand" "=v")
(float_extend:V2DF
(vec_select:V2SF
- (match_operand:V4SF 1 "nonimmediate_operand" "vm")
+ (match_operand:V4SF 1 "vector_operand" "vBm")
(parallel [(const_int 0) (const_int 1)]))))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
"%vcvtps2pd\t{%1, %0<mask_operand2>|%0<mask_operand2>, %q1}"
@@ -5495,7 +5495,7 @@
(vec_select:V4SF
(vec_concat:V8SF
(match_dup 2)
- (match_operand:V4SF 1 "nonimmediate_operand"))
+ (match_operand:V4SF 1 "vector_operand"))
(parallel [(const_int 6) (const_int 7)
(const_int 2) (const_int 3)])))
(set (match_operand:V2DF 0 "register_operand")
@@ -5536,7 +5536,7 @@
[(set (match_operand:V2DF 0 "register_operand")
(float_extend:V2DF
(vec_select:V2SF
- (match_operand:V4SF 1 "nonimmediate_operand")
+ (match_operand:V4SF 1 "vector_operand")
(parallel [(const_int 0) (const_int 1)]))))]
"TARGET_SSE2")
@@ -5608,7 +5608,7 @@
(define_expand "vec_unpacks_float_hi_v4si"
[(set (match_dup 2)
(vec_select:V4SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 2) (const_int 3)
(const_int 2) (const_int 3)])))
(set (match_operand:V2DF 0 "register_operand")
@@ -5623,14 +5623,14 @@
[(set (match_operand:V2DF 0 "register_operand")
(float:V2DF
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 0) (const_int 1)]))))]
"TARGET_SSE2")
(define_expand "vec_unpacks_float_hi_v8si"
[(set (match_dup 2)
(vec_select:V4SI
- (match_operand:V8SI 1 "nonimmediate_operand")
+ (match_operand:V8SI 1 "vector_operand")
(parallel [(const_int 4) (const_int 5)
(const_int 6) (const_int 7)])))
(set (match_operand:V4DF 0 "register_operand")
@@ -5676,7 +5676,7 @@
(define_expand "vec_unpacku_float_hi_v4si"
[(set (match_dup 5)
(vec_select:V4SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 2) (const_int 3)
(const_int 2) (const_int 3)])))
(set (match_dup 6)
@@ -5713,7 +5713,7 @@
[(set (match_dup 5)
(float:V2DF
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 0) (const_int 1)]))))
(set (match_dup 6)
(lt:V2DF (match_dup 5) (match_dup 3)))
@@ -5855,8 +5855,8 @@
(define_expand "vec_pack_trunc_v2df"
[(match_operand:V4SF 0 "register_operand")
- (match_operand:V2DF 1 "nonimmediate_operand")
- (match_operand:V2DF 2 "nonimmediate_operand")]
+ (match_operand:V2DF 1 "vector_operand")
+ (match_operand:V2DF 2 "vector_operand")]
"TARGET_SSE2"
{
rtx tmp0, tmp1;
@@ -5917,8 +5917,8 @@
(define_expand "vec_pack_sfix_trunc_v2df"
[(match_operand:V4SI 0 "register_operand")
- (match_operand:V2DF 1 "nonimmediate_operand")
- (match_operand:V2DF 2 "nonimmediate_operand")]
+ (match_operand:V2DF 1 "vector_operand")
+ (match_operand:V2DF 2 "vector_operand")]
"TARGET_SSE2"
{
rtx tmp0, tmp1, tmp2;
@@ -6014,8 +6014,8 @@
(define_expand "vec_pack_sfix_v2df"
[(match_operand:V4SI 0 "register_operand")
- (match_operand:V2DF 1 "nonimmediate_operand")
- (match_operand:V2DF 2 "nonimmediate_operand")]
+ (match_operand:V2DF 1 "vector_operand")
+ (match_operand:V2DF 2 "vector_operand")]
"TARGET_SSE2"
{
rtx tmp0, tmp1, tmp2;
@@ -6218,7 +6218,7 @@
(vec_select:V4SF
(vec_concat:V8SF
(match_operand:V4SF 1 "register_operand" "0,v")
- (match_operand:V4SF 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V4SF 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 2) (const_int 6)
(const_int 3) (const_int 7)])))]
"TARGET_SSE && <mask_avx512vl_condition>"
@@ -6323,7 +6323,7 @@
(vec_select:V4SF
(vec_concat:V8SF
(match_operand:V4SF 1 "register_operand" "0,x")
- (match_operand:V4SF 2 "nonimmediate_operand" "xm,xm"))
+ (match_operand:V4SF 2 "vector_operand" "xBm,xm"))
(parallel [(const_int 0) (const_int 4)
(const_int 1) (const_int 5)])))]
"TARGET_SSE"
@@ -6357,7 +6357,7 @@
[(set (match_operand:V4SF 0 "register_operand" "=v")
(vec_select:V4SF
(vec_concat:V8SF
- (match_operand:V4SF 1 "nonimmediate_operand" "vm")
+ (match_operand:V4SF 1 "vector_operand" "vBm")
(match_dup 1))
(parallel [(const_int 1)
(const_int 1)
@@ -6410,7 +6410,7 @@
[(set (match_operand:V4SF 0 "register_operand" "=v")
(vec_select:V4SF
(vec_concat:V8SF
- (match_operand:V4SF 1 "nonimmediate_operand" "vm")
+ (match_operand:V4SF 1 "vector_operand" "vBm")
(match_dup 1))
(parallel [(const_int 0)
(const_int 0)
@@ -6505,7 +6505,7 @@
(define_expand "sse_shufps<mask_expand4_name>"
[(match_operand:V4SF 0 "register_operand")
(match_operand:V4SF 1 "register_operand")
- (match_operand:V4SF 2 "nonimmediate_operand")
+ (match_operand:V4SF 2 "vector_operand")
(match_operand:SI 3 "const_int_operand")]
"TARGET_SSE"
{
@@ -6555,7 +6555,7 @@
(vec_select:VI4F_128
(vec_concat:<ssedoublevecmode>
(match_operand:VI4F_128 1 "register_operand" "0,x")
- (match_operand:VI4F_128 2 "nonimmediate_operand" "xm,xm"))
+ (match_operand:VI4F_128 2 "vector_operand" "xBm,xm"))
(parallel [(match_operand 3 "const_0_to_3_operand")
(match_operand 4 "const_0_to_3_operand")
(match_operand 5 "const_4_to_7_operand")
@@ -6651,6 +6651,7 @@
%vmovaps\t{%1, %0|%0, %1}
%vmovlps\t{%1, %d0|%d0, %q1}"
[(set_attr "type" "ssemov")
+ (set_attr "ssememalign" "64")
(set_attr "prefix" "maybe_vex")
(set_attr "mode" "V2SF,V4SF,V2SF")])
@@ -6776,7 +6777,7 @@
(set_attr "mode" "V4SF,V4SF,V4SF,V4SF,V4SF,V4SF,SF,DI,DI")])
;; ??? In theory we can match memory for the MMX alternative, but allowing
-;; nonimmediate_operand for operand 2 and *not* allowing memory for the SSE
+;; vector_operand for operand 2 and *not* allowing memory for the SSE
;; alternatives pretty much forces the MMX alternative to be chosen.
(define_insn "*vec_concatv2sf_sse"
[(set (match_operand:V2SF 0 "register_operand" "=x,x,*y,*y")
@@ -8501,7 +8502,7 @@
(define_expand "sse2_shufpd<mask_expand4_name>"
[(match_operand:V2DF 0 "register_operand")
(match_operand:V2DF 1 "register_operand")
- (match_operand:V2DF 2 "nonimmediate_operand")
+ (match_operand:V2DF 2 "vector_operand")
(match_operand:SI 3 "const_int_operand")]
"TARGET_SSE2"
{
@@ -8576,7 +8577,7 @@
(vec_select:V2DI
(vec_concat:V4DI
(match_operand:V2DI 1 "register_operand" "0,v")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V2DI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 1)
(const_int 3)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
@@ -8626,7 +8627,7 @@
(vec_select:V2DI
(vec_concat:V4DI
(match_operand:V2DI 1 "register_operand" "0,v")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V2DI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 0)
(const_int 2)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
@@ -8644,7 +8645,7 @@
(vec_select:VI8F_128
(vec_concat:<ssedoublevecmode>
(match_operand:VI8F_128 1 "register_operand" "0,x")
- (match_operand:VI8F_128 2 "nonimmediate_operand" "xm,xm"))
+ (match_operand:VI8F_128 2 "vector_operand" "xBm,xm"))
(parallel [(match_operand 3 "const_0_to_1_operand")
(match_operand 4 "const_2_to_3_operand")])))]
"TARGET_SSE2"
@@ -9615,15 +9616,15 @@
[(set (match_operand:VI_AVX2 0 "register_operand")
(minus:VI_AVX2
(match_dup 2)
- (match_operand:VI_AVX2 1 "nonimmediate_operand")))]
+ (match_operand:VI_AVX2 1 "vector_operand")))]
"TARGET_SSE2"
"operands[2] = force_reg (<MODE>mode, CONST0_RTX (<MODE>mode));")
(define_expand "<plusminus_insn><mode>3"
[(set (match_operand:VI_AVX2 0 "register_operand")
(plusminus:VI_AVX2
- (match_operand:VI_AVX2 1 "nonimmediate_operand")
- (match_operand:VI_AVX2 2 "nonimmediate_operand")))]
+ (match_operand:VI_AVX2 1 "vector_operand")
+ (match_operand:VI_AVX2 2 "vector_operand")))]
"TARGET_SSE2"
"ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
@@ -9652,8 +9653,8 @@
(define_insn "*<plusminus_insn><mode>3"
[(set (match_operand:VI_AVX2 0 "register_operand" "=x,v")
(plusminus:VI_AVX2
- (match_operand:VI_AVX2 1 "nonimmediate_operand" "<comm>0,v")
- (match_operand:VI_AVX2 2 "nonimmediate_operand" "xBm,vm")))]
+ (match_operand:VI_AVX2 1 "vector_operand" "<comm>0,v")
+ (match_operand:VI_AVX2 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE2
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
"@
@@ -9697,16 +9698,16 @@
(define_expand "<sse2_avx2>_<plusminus_insn><mode>3<mask_name>"
[(set (match_operand:VI12_AVX2 0 "register_operand")
(sat_plusminus:VI12_AVX2
- (match_operand:VI12_AVX2 1 "nonimmediate_operand")
- (match_operand:VI12_AVX2 2 "nonimmediate_operand")))]
+ (match_operand:VI12_AVX2 1 "vector_operand")
+ (match_operand:VI12_AVX2 2 "vector_operand")))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
(define_insn "*<sse2_avx2>_<plusminus_insn><mode>3<mask_name>"
[(set (match_operand:VI12_AVX2 0 "register_operand" "=x,v")
(sat_plusminus:VI12_AVX2
- (match_operand:VI12_AVX2 1 "nonimmediate_operand" "<comm>0,v")
- (match_operand:VI12_AVX2 2 "nonimmediate_operand" "xBm,vm")))]
+ (match_operand:VI12_AVX2 1 "vector_operand" "<comm>0,v")
+ (match_operand:VI12_AVX2 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
"@
@@ -9730,15 +9731,15 @@
(define_expand "mul<mode>3<mask_name>"
[(set (match_operand:VI2_AVX2 0 "register_operand")
- (mult:VI2_AVX2 (match_operand:VI2_AVX2 1 "nonimmediate_operand")
- (match_operand:VI2_AVX2 2 "nonimmediate_operand")))]
+ (mult:VI2_AVX2 (match_operand:VI2_AVX2 1 "vector_operand")
+ (match_operand:VI2_AVX2 2 "vector_operand")))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"ix86_fixup_binary_operands_no_copy (MULT, <MODE>mode, operands);")
(define_insn "*mul<mode>3<mask_name>"
[(set (match_operand:VI2_AVX2 0 "register_operand" "=x,v")
- (mult:VI2_AVX2 (match_operand:VI2_AVX2 1 "nonimmediate_operand" "%0,v")
- (match_operand:VI2_AVX2 2 "nonimmediate_operand" "xm,vm")))]
+ (mult:VI2_AVX2 (match_operand:VI2_AVX2 1 "vector_operand" "%0,v")
+ (match_operand:VI2_AVX2 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE2
&& ix86_binary_operator_ok (MULT, <MODE>mode, operands)
&& <mask_mode512bit_condition> && <mask_avx512bw_condition>"
@@ -9757,9 +9758,9 @@
(lshiftrt:<ssedoublemode>
(mult:<ssedoublemode>
(any_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 1 "nonimmediate_operand"))
+ (match_operand:VI2_AVX2 1 "vector_operand"))
(any_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 2 "nonimmediate_operand")))
+ (match_operand:VI2_AVX2 2 "vector_operand")))
(const_int 16))))]
"TARGET_SSE2
&& <mask_mode512bit_condition> && <mask_avx512bw_condition>"
@@ -9771,9 +9772,9 @@
(lshiftrt:<ssedoublemode>
(mult:<ssedoublemode>
(any_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 1 "nonimmediate_operand" "%0,v"))
+ (match_operand:VI2_AVX2 1 "vector_operand" "%0,v"))
(any_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 2 "nonimmediate_operand" "xm,vm")))
+ (match_operand:VI2_AVX2 2 "vector_operand" "xBm,vm")))
(const_int 16))))]
"TARGET_SSE2
&& ix86_binary_operator_ok (MULT, <MODE>mode, operands)
@@ -9873,11 +9874,11 @@
(mult:V2DI
(zero_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 0) (const_int 2)])))
(zero_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 2 "nonimmediate_operand")
+ (match_operand:V4SI 2 "vector_operand")
(parallel [(const_int 0) (const_int 2)])))))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
"ix86_fixup_binary_operands_no_copy (MULT, V4SImode, operands);")
@@ -9887,11 +9888,11 @@
(mult:V2DI
(zero_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand" "%0,v")
+ (match_operand:V4SI 1 "vector_operand" "%0,v")
(parallel [(const_int 0) (const_int 2)])))
(zero_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 2 "nonimmediate_operand" "xm,vm")
+ (match_operand:V4SI 2 "vector_operand" "xBm,vm")
(parallel [(const_int 0) (const_int 2)])))))]
"TARGET_SSE2 && <mask_avx512vl_condition>
&& ix86_binary_operator_ok (MULT, V4SImode, operands)"
@@ -9991,11 +9992,11 @@
(mult:V2DI
(sign_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(parallel [(const_int 0) (const_int 2)])))
(sign_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 2 "nonimmediate_operand")
+ (match_operand:V4SI 2 "vector_operand")
(parallel [(const_int 0) (const_int 2)])))))]
"TARGET_SSE4_1 && <mask_avx512vl_condition>"
"ix86_fixup_binary_operands_no_copy (MULT, V4SImode, operands);")
@@ -10005,11 +10006,11 @@
(mult:V2DI
(sign_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 1 "nonimmediate_operand" "%0,0,v")
+ (match_operand:V4SI 1 "vector_operand" "%0,0,v")
(parallel [(const_int 0) (const_int 2)])))
(sign_extend:V2DI
(vec_select:V2SI
- (match_operand:V4SI 2 "nonimmediate_operand" "Yrm,*xm,vm")
+ (match_operand:V4SI 2 "vector_operand" "YrBm,*xBm,vm")
(parallel [(const_int 0) (const_int 2)])))))]
"TARGET_SSE4_1 && <mask_avx512vl_condition>
&& ix86_binary_operator_ok (MULT, V4SImode, operands)"
@@ -10113,12 +10114,12 @@
(mult:V4SI
(sign_extend:V4SI
(vec_select:V4HI
- (match_operand:V8HI 1 "nonimmediate_operand")
+ (match_operand:V8HI 1 "vector_operand")
(parallel [(const_int 0) (const_int 2)
(const_int 4) (const_int 6)])))
(sign_extend:V4SI
(vec_select:V4HI
- (match_operand:V8HI 2 "nonimmediate_operand")
+ (match_operand:V8HI 2 "vector_operand")
(parallel [(const_int 0) (const_int 2)
(const_int 4) (const_int 6)]))))
(mult:V4SI
@@ -10139,12 +10140,12 @@
(mult:V4SI
(sign_extend:V4SI
(vec_select:V4HI
- (match_operand:V8HI 1 "nonimmediate_operand" "%0,x")
+ (match_operand:V8HI 1 "vector_operand" "%0,x")
(parallel [(const_int 0) (const_int 2)
(const_int 4) (const_int 6)])))
(sign_extend:V4SI
(vec_select:V4HI
- (match_operand:V8HI 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V8HI 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0) (const_int 2)
(const_int 4) (const_int 6)]))))
(mult:V4SI
@@ -10187,9 +10188,9 @@
{
if (TARGET_SSE4_1)
{
- if (!nonimmediate_operand (operands[1], <MODE>mode))
+ if (!vector_operand (operands[1], <MODE>mode))
operands[1] = force_reg (<MODE>mode, operands[1]);
- if (!nonimmediate_operand (operands[2], <MODE>mode))
+ if (!vector_operand (operands[2], <MODE>mode))
operands[2] = force_reg (<MODE>mode, operands[2]);
ix86_fixup_binary_operands_no_copy (MULT, <MODE>mode, operands);
}
@@ -10203,8 +10204,8 @@
(define_insn "*<sse4_1_avx2>_mul<mode>3<mask_name>"
[(set (match_operand:VI4_AVX512F 0 "register_operand" "=Yr,*x,v")
(mult:VI4_AVX512F
- (match_operand:VI4_AVX512F 1 "nonimmediate_operand" "%0,0,v")
- (match_operand:VI4_AVX512F 2 "nonimmediate_operand" "Yrm,*xm,vm")))]
+ (match_operand:VI4_AVX512F 1 "vector_operand" "%0,0,v")
+ (match_operand:VI4_AVX512F 2 "vector_operand" "YrBm,*xBm,vm")))]
"TARGET_SSE4_1 && ix86_binary_operator_ok (MULT, <MODE>mode, operands) && <mask_mode512bit_condition>"
"@
pmulld\t{%2, %0|%0, %2}
@@ -10256,8 +10257,8 @@
;; named patterns, but signed V4SI needs special help for plain SSE2.
(define_expand "vec_widen_smult_even_v4si"
[(match_operand:V2DI 0 "register_operand")
- (match_operand:V4SI 1 "nonimmediate_operand")
- (match_operand:V4SI 2 "nonimmediate_operand")]
+ (match_operand:V4SI 1 "vector_operand")
+ (match_operand:V4SI 2 "vector_operand")]
"TARGET_SSE2"
{
ix86_expand_mul_widen_evenodd (operands[0], operands[1], operands[2],
@@ -10313,8 +10314,8 @@
(define_expand "usadv16qi"
[(match_operand:V4SI 0 "register_operand")
(match_operand:V16QI 1 "register_operand")
- (match_operand:V16QI 2 "nonimmediate_operand")
- (match_operand:V4SI 3 "nonimmediate_operand")]
+ (match_operand:V16QI 2 "vector_operand")
+ (match_operand:V4SI 3 "vector_operand")]
"TARGET_SSE2"
{
rtx t1 = gen_reg_rtx (V2DImode);
@@ -10654,8 +10655,8 @@
(define_expand "<code><mode>3"
[(set (match_operand:VI124_128 0 "register_operand")
(smaxmin:VI124_128
- (match_operand:VI124_128 1 "nonimmediate_operand")
- (match_operand:VI124_128 2 "nonimmediate_operand")))]
+ (match_operand:VI124_128 1 "vector_operand")
+ (match_operand:VI124_128 2 "vector_operand")))]
"TARGET_SSE2"
{
if (TARGET_SSE4_1 || <MODE>mode == V8HImode)
@@ -10693,8 +10694,8 @@
(define_insn "*sse4_1_<code><mode>3<mask_name>"
[(set (match_operand:VI14_128 0 "register_operand" "=Yr,*x,v")
(smaxmin:VI14_128
- (match_operand:VI14_128 1 "nonimmediate_operand" "%0,0,v")
- (match_operand:VI14_128 2 "nonimmediate_operand" "Yrm,*xm,vm")))]
+ (match_operand:VI14_128 1 "vector_operand" "%0,0,v")
+ (match_operand:VI14_128 2 "vector_operand" "YrBm,*xBm,vm")))]
"TARGET_SSE4_1
&& <mask_mode512bit_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
@@ -10711,8 +10712,8 @@
(define_insn "*<code>v8hi3"
[(set (match_operand:V8HI 0 "register_operand" "=x,x")
(smaxmin:V8HI
- (match_operand:V8HI 1 "nonimmediate_operand" "%0,x")
- (match_operand:V8HI 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:V8HI 1 "vector_operand" "%0,x")
+ (match_operand:V8HI 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE2 && ix86_binary_operator_ok (<CODE>, V8HImode, operands)"
"@
p<maxmin_int>w\t{%2, %0|%0, %2}
@@ -10727,8 +10728,8 @@
(define_expand "<code><mode>3"
[(set (match_operand:VI124_128 0 "register_operand")
(umaxmin:VI124_128
- (match_operand:VI124_128 1 "nonimmediate_operand")
- (match_operand:VI124_128 2 "nonimmediate_operand")))]
+ (match_operand:VI124_128 1 "vector_operand")
+ (match_operand:VI124_128 2 "vector_operand")))]
"TARGET_SSE2"
{
if (TARGET_SSE4_1 || <MODE>mode == V16QImode)
@@ -10777,8 +10778,8 @@
(define_insn "*sse4_1_<code><mode>3<mask_name>"
[(set (match_operand:VI24_128 0 "register_operand" "=Yr,*x,v")
(umaxmin:VI24_128
- (match_operand:VI24_128 1 "nonimmediate_operand" "%0,0,v")
- (match_operand:VI24_128 2 "nonimmediate_operand" "Yrm,*xm,vm")))]
+ (match_operand:VI24_128 1 "vector_operand" "%0,0,v")
+ (match_operand:VI24_128 2 "vector_operand" "YrBm,*xBm,vm")))]
"TARGET_SSE4_1
&& <mask_mode512bit_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
@@ -10795,8 +10796,8 @@
(define_insn "*<code>v16qi3"
[(set (match_operand:V16QI 0 "register_operand" "=x,x")
(umaxmin:V16QI
- (match_operand:V16QI 1 "nonimmediate_operand" "%0,x")
- (match_operand:V16QI 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:V16QI 1 "vector_operand" "%0,x")
+ (match_operand:V16QI 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE2 && ix86_binary_operator_ok (<CODE>, V16QImode, operands)"
"@
p<maxmin_int>b\t{%2, %0|%0, %2}
@@ -10881,8 +10882,8 @@
(define_insn "*sse4_1_eqv2di3"
[(set (match_operand:V2DI 0 "register_operand" "=Yr,*x,x")
(eq:V2DI
- (match_operand:V2DI 1 "nonimmediate_operand" "%0,0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "Yrm,*xm,xm")))]
+ (match_operand:V2DI 1 "vector_operand" "%0,0,x")
+ (match_operand:V2DI 2 "vector_operand" "YrBm,*xBm,xm")))]
"TARGET_SSE4_1 && ix86_binary_operator_ok (EQ, V2DImode, operands)"
"@
pcmpeqq\t{%2, %0|%0, %2}
@@ -10897,8 +10898,8 @@
(define_insn "*sse2_eq<mode>3"
[(set (match_operand:VI124_128 0 "register_operand" "=x,x")
(eq:VI124_128
- (match_operand:VI124_128 1 "nonimmediate_operand" "%0,x")
- (match_operand:VI124_128 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:VI124_128 1 "vector_operand" "%0,x")
+ (match_operand:VI124_128 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE2 && !TARGET_XOP
&& ix86_binary_operator_ok (EQ, <MODE>mode, operands)"
"@
@@ -10913,16 +10914,16 @@
(define_expand "sse2_eq<mode>3"
[(set (match_operand:VI124_128 0 "register_operand")
(eq:VI124_128
- (match_operand:VI124_128 1 "nonimmediate_operand")
- (match_operand:VI124_128 2 "nonimmediate_operand")))]
+ (match_operand:VI124_128 1 "vector_operand")
+ (match_operand:VI124_128 2 "vector_operand")))]
"TARGET_SSE2 && !TARGET_XOP "
"ix86_fixup_binary_operands_no_copy (EQ, <MODE>mode, operands);")
(define_expand "sse4_1_eqv2di3"
[(set (match_operand:V2DI 0 "register_operand")
(eq:V2DI
- (match_operand:V2DI 1 "nonimmediate_operand")
- (match_operand:V2DI 2 "nonimmediate_operand")))]
+ (match_operand:V2DI 1 "vector_operand")
+ (match_operand:V2DI 2 "vector_operand")))]
"TARGET_SSE4_1"
"ix86_fixup_binary_operands_no_copy (EQ, V2DImode, operands);")
@@ -10930,7 +10931,7 @@
[(set (match_operand:V2DI 0 "register_operand" "=Yr,*x,x")
(gt:V2DI
(match_operand:V2DI 1 "register_operand" "0,0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "Yrm,*xm,xm")))]
+ (match_operand:V2DI 2 "vector_operand" "YrBm,*xBm,xm")))]
"TARGET_SSE4_2"
"@
pcmpgtq\t{%2, %0|%0, %2}
@@ -10982,7 +10983,7 @@
[(set (match_operand:VI124_128 0 "register_operand" "=x,x")
(gt:VI124_128
(match_operand:VI124_128 1 "register_operand" "0,x")
- (match_operand:VI124_128 2 "nonimmediate_operand" "xm,xm")))]
+ (match_operand:VI124_128 2 "vector_operand" "xBm,xm")))]
"TARGET_SSE2 && !TARGET_XOP"
"@
pcmpgt<ssemodesuffix>\t{%2, %0|%0, %2}
@@ -11031,7 +11032,7 @@
[(set (match_operand:V_128 0 "register_operand")
(if_then_else:V_128
(match_operator 3 ""
- [(match_operand:VI124_128 4 "nonimmediate_operand")
+ [(match_operand:VI124_128 4 "vector_operand")
(match_operand:VI124_128 5 "general_operand")])
(match_operand:V_128 1)
(match_operand:V_128 2)))]
@@ -11048,7 +11049,7 @@
[(set (match_operand:VI8F_128 0 "register_operand")
(if_then_else:VI8F_128
(match_operator 3 ""
- [(match_operand:V2DI 4 "nonimmediate_operand")
+ [(match_operand:V2DI 4 "vector_operand")
(match_operand:V2DI 5 "general_operand")])
(match_operand:VI8F_128 1)
(match_operand:VI8F_128 2)))]
@@ -11097,8 +11098,8 @@
[(set (match_operand:V_128 0 "register_operand")
(if_then_else:V_128
(match_operator 3 ""
- [(match_operand:VI124_128 4 "nonimmediate_operand")
- (match_operand:VI124_128 5 "nonimmediate_operand")])
+ [(match_operand:VI124_128 4 "vector_operand")
+ (match_operand:VI124_128 5 "vector_operand")])
(match_operand:V_128 1 "general_operand")
(match_operand:V_128 2 "general_operand")))]
"TARGET_SSE2
@@ -11114,8 +11115,8 @@
[(set (match_operand:VI8F_128 0 "register_operand")
(if_then_else:VI8F_128
(match_operator 3 ""
- [(match_operand:V2DI 4 "nonimmediate_operand")
- (match_operand:V2DI 5 "nonimmediate_operand")])
+ [(match_operand:V2DI 4 "vector_operand")
+ (match_operand:V2DI 5 "vector_operand")])
(match_operand:VI8F_128 1 "general_operand")
(match_operand:VI8F_128 2 "general_operand")))]
"TARGET_SSE4_2"
@@ -11177,7 +11178,7 @@
(define_expand "one_cmpl<mode>2"
[(set (match_operand:VI 0 "register_operand")
- (xor:VI (match_operand:VI 1 "nonimmediate_operand")
+ (xor:VI (match_operand:VI 1 "vector_operand")
(match_dup 2)))]
"TARGET_SSE"
{
@@ -11194,7 +11195,7 @@
[(set (match_operand:VI_AVX2 0 "register_operand")
(and:VI_AVX2
(not:VI_AVX2 (match_operand:VI_AVX2 1 "register_operand"))
- (match_operand:VI_AVX2 2 "nonimmediate_operand")))]
+ (match_operand:VI_AVX2 2 "vector_operand")))]
"TARGET_SSE2")
(define_expand "<sse2_avx2>_andnot<mode>3_mask"
@@ -11223,7 +11224,7 @@
[(set (match_operand:VI 0 "register_operand" "=x,v")
(and:VI
(not:VI (match_operand:VI 1 "register_operand" "0,v"))
- (match_operand:VI 2 "nonimmediate_operand" "xm,vm")))]
+ (match_operand:VI 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE"
{
static char buf[64];
@@ -11360,8 +11361,8 @@
(define_insn "<mask_codefor><code><mode>3<mask_name>"
[(set (match_operand:VI48_AVX_AVX512F 0 "register_operand" "=x,v")
(any_logic:VI48_AVX_AVX512F
- (match_operand:VI48_AVX_AVX512F 1 "nonimmediate_operand" "%0,v")
- (match_operand:VI48_AVX_AVX512F 2 "nonimmediate_operand" "xBm,vm")))]
+ (match_operand:VI48_AVX_AVX512F 1 "vector_operand" "%0,v")
+ (match_operand:VI48_AVX_AVX512F 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE && <mask_mode512bit_condition>
&& ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
@@ -11456,8 +11457,8 @@
(define_insn "*<code><mode>3"
[(set (match_operand:VI12_AVX_AVX512F 0 "register_operand" "=x,v")
(any_logic: VI12_AVX_AVX512F
- (match_operand:VI12_AVX_AVX512F 1 "nonimmediate_operand" "%0,v")
- (match_operand:VI12_AVX_AVX512F 2 "nonimmediate_operand" "xBm,vm")))]
+ (match_operand:VI12_AVX_AVX512F 1 "vector_operand" "%0,v")
+ (match_operand:VI12_AVX_AVX512F 2 "vector_operand" "xBm,vm")))]
"TARGET_SSE && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
static char buf[64];
@@ -11637,7 +11638,7 @@
(ss_truncate:<ssehalfvecmode>
(match_operand:<sseunpackmode> 1 "register_operand" "0,v"))
(ss_truncate:<ssehalfvecmode>
- (match_operand:<sseunpackmode> 2 "nonimmediate_operand" "xm,vm"))))]
+ (match_operand:<sseunpackmode> 2 "vector_operand" "xBm,vm"))))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"@
packsswb\t{%2, %0|%0, %2}
@@ -11654,7 +11655,7 @@
(ss_truncate:<ssehalfvecmode>
(match_operand:<sseunpackmode> 1 "register_operand" "0,v"))
(ss_truncate:<ssehalfvecmode>
- (match_operand:<sseunpackmode> 2 "nonimmediate_operand" "xm,vm"))))]
+ (match_operand:<sseunpackmode> 2 "vector_operand" "xBm,vm"))))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"@
packssdw\t{%2, %0|%0, %2}
@@ -11671,7 +11672,7 @@
(us_truncate:<ssehalfvecmode>
(match_operand:<sseunpackmode> 1 "register_operand" "0,v"))
(us_truncate:<ssehalfvecmode>
- (match_operand:<sseunpackmode> 2 "nonimmediate_operand" "xm,vm"))))]
+ (match_operand:<sseunpackmode> 2 "vector_operand" "xBm,vm"))))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"@
packuswb\t{%2, %0|%0, %2}
@@ -11759,7 +11760,7 @@
(vec_select:V16QI
(vec_concat:V32QI
(match_operand:V16QI 1 "register_operand" "0,v")
- (match_operand:V16QI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V16QI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 8) (const_int 24)
(const_int 9) (const_int 25)
(const_int 10) (const_int 26)
@@ -11855,7 +11856,7 @@
(vec_select:V16QI
(vec_concat:V32QI
(match_operand:V16QI 1 "register_operand" "0,v")
- (match_operand:V16QI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V16QI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 0) (const_int 16)
(const_int 1) (const_int 17)
(const_int 2) (const_int 18)
@@ -11927,7 +11928,7 @@
(vec_select:V8HI
(vec_concat:V16HI
(match_operand:V8HI 1 "register_operand" "0,v")
- (match_operand:V8HI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V8HI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 4) (const_int 12)
(const_int 5) (const_int 13)
(const_int 6) (const_int 14)
@@ -11995,7 +11996,7 @@
(vec_select:V8HI
(vec_concat:V16HI
(match_operand:V8HI 1 "register_operand" "0,v")
- (match_operand:V8HI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V8HI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 0) (const_int 8)
(const_int 1) (const_int 9)
(const_int 2) (const_int 10)
@@ -12052,7 +12053,7 @@
(vec_select:V4SI
(vec_concat:V8SI
(match_operand:V4SI 1 "register_operand" "0,v")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V4SI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 2) (const_int 6)
(const_int 3) (const_int 7)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
@@ -12106,7 +12107,7 @@
(vec_select:V4SI
(vec_concat:V8SI
(match_operand:V4SI 1 "register_operand" "0,v")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm,vm"))
+ (match_operand:V4SI 2 "vector_operand" "xBm,vm"))
(parallel [(const_int 0) (const_int 4)
(const_int 1) (const_int 5)])))]
"TARGET_SSE2 && <mask_avx512vl_condition>"
@@ -12773,7 +12774,7 @@
(define_expand "sse2_pshufd"
[(match_operand:V4SI 0 "register_operand")
- (match_operand:V4SI 1 "nonimmediate_operand")
+ (match_operand:V4SI 1 "vector_operand")
(match_operand:SI 2 "const_int_operand")]
"TARGET_SSE2"
{
@@ -12789,7 +12790,7 @@
(define_insn "sse2_pshufd_1<mask_name>"
[(set (match_operand:V4SI 0 "register_operand" "=v")
(vec_select:V4SI
- (match_operand:V4SI 1 "nonimmediate_operand" "vm")
+ (match_operand:V4SI 1 "vector_operand" "vBm")
(parallel [(match_operand 2 "const_0_to_3_operand")
(match_operand 3 "const_0_to_3_operand")
(match_operand 4 "const_0_to_3_operand")
@@ -12925,7 +12926,7 @@
(define_expand "sse2_pshuflw"
[(match_operand:V8HI 0 "register_operand")
- (match_operand:V8HI 1 "nonimmediate_operand")
+ (match_operand:V8HI 1 "vector_operand")
(match_operand:SI 2 "const_int_operand")]
"TARGET_SSE2"
{
@@ -12941,7 +12942,7 @@
(define_insn "sse2_pshuflw_1<mask_name>"
[(set (match_operand:V8HI 0 "register_operand" "=v")
(vec_select:V8HI
- (match_operand:V8HI 1 "nonimmediate_operand" "vm")
+ (match_operand:V8HI 1 "vector_operand" "vBm")
(parallel [(match_operand 2 "const_0_to_3_operand")
(match_operand 3 "const_0_to_3_operand")
(match_operand 4 "const_0_to_3_operand")
@@ -13082,7 +13083,7 @@
(define_expand "sse2_pshufhw"
[(match_operand:V8HI 0 "register_operand")
- (match_operand:V8HI 1 "nonimmediate_operand")
+ (match_operand:V8HI 1 "vector_operand")
(match_operand:SI 2 "const_int_operand")]
"TARGET_SSE2"
{
@@ -13098,7 +13099,7 @@
(define_insn "sse2_pshufhw_1<mask_name>"
[(set (match_operand:V8HI 0 "register_operand" "=v")
(vec_select:V8HI
- (match_operand:V8HI 1 "nonimmediate_operand" "vm")
+ (match_operand:V8HI 1 "vector_operand" "vBm")
(parallel [(const_int 0)
(const_int 1)
(const_int 2)
@@ -13576,9 +13577,9 @@
(plus:<ssedoublemode>
(plus:<ssedoublemode>
(zero_extend:<ssedoublemode>
- (match_operand:VI12_AVX2 1 "nonimmediate_operand"))
+ (match_operand:VI12_AVX2 1 "vector_operand"))
(zero_extend:<ssedoublemode>
- (match_operand:VI12_AVX2 2 "nonimmediate_operand")))
+ (match_operand:VI12_AVX2 2 "vector_operand")))
(match_dup <mask_expand_op3>))
(const_int 1))))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
@@ -13603,9 +13604,9 @@
(plus:<ssedoublemode>
(plus:<ssedoublemode>
(zero_extend:<ssedoublemode>
- (match_operand:VI12_AVX2 1 "nonimmediate_operand" "%0,v"))
+ (match_operand:VI12_AVX2 1 "vector_operand" "%0,v"))
(zero_extend:<ssedoublemode>
- (match_operand:VI12_AVX2 2 "nonimmediate_operand" "xm,vm")))
+ (match_operand:VI12_AVX2 2 "vector_operand" "xBm,vm")))
(match_operand:VI12_AVX2 <mask_expand_op3> "const1_operand"))
(const_int 1))))]
"TARGET_SSE2 && <mask_mode512bit_condition> && <mask_avx512bw_condition>
@@ -13625,7 +13626,7 @@
[(set (match_operand:VI8_AVX2_AVX512BW 0 "register_operand" "=x,v")
(unspec:VI8_AVX2_AVX512BW
[(match_operand:<ssebytemode> 1 "register_operand" "0,v")
- (match_operand:<ssebytemode> 2 "nonimmediate_operand" "xm,vm")]
+ (match_operand:<ssebytemode> 2 "vector_operand" "xBm,vm")]
UNSPEC_PSADBW))]
"TARGET_SSE2"
"@
@@ -13891,7 +13892,7 @@
(vec_concat:V2HI
(ssse3_plusminus:HI
(vec_select:HI
- (match_operand:V8HI 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V8HI 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0)]))
(vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
(ssse3_plusminus:HI
@@ -14004,7 +14005,7 @@
(vec_concat:V2SI
(plusminus:SI
(vec_select:SI
- (match_operand:V4SI 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V4SI 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0)]))
(vec_select:SI (match_dup 2) (parallel [(const_int 1)])))
(plusminus:SI
@@ -14159,7 +14160,7 @@
(const_int 12) (const_int 14)])))
(sign_extend:V8HI
(vec_select:V8QI
- (match_operand:V16QI 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V16QI 2 "vector_operand" "xBm,xm")
(parallel [(const_int 0) (const_int 2)
(const_int 4) (const_int 6)
(const_int 8) (const_int 10)
@@ -14274,9 +14275,9 @@
(lshiftrt:<ssedoublemode>
(mult:<ssedoublemode>
(sign_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 1 "nonimmediate_operand" "%0,v"))
+ (match_operand:VI2_AVX2 1 "vector_operand" "%0,v"))
(sign_extend:<ssedoublemode>
- (match_operand:VI2_AVX2 2 "nonimmediate_operand" "xm,vm")))
+ (match_operand:VI2_AVX2 2 "vector_operand" "xBm,vm")))
(const_int 14))
(match_operand:VI2_AVX2 3 "const1_operand"))
(const_int 1))))]
@@ -14317,7 +14318,7 @@
[(set (match_operand:VI1_AVX512 0 "register_operand" "=x,v")
(unspec:VI1_AVX512
[(match_operand:VI1_AVX512 1 "register_operand" "0,v")
- (match_operand:VI1_AVX512 2 "nonimmediate_operand" "xm,vm")]
+ (match_operand:VI1_AVX512 2 "vector_operand" "xBm,vm")]
UNSPEC_PSHUFB))]
"TARGET_SSSE3 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"@
@@ -14347,7 +14348,7 @@
[(set (match_operand:VI124_AVX2 0 "register_operand" "=x,x")
(unspec:VI124_AVX2
[(match_operand:VI124_AVX2 1 "register_operand" "0,x")
- (match_operand:VI124_AVX2 2 "nonimmediate_operand" "xm,xm")]
+ (match_operand:VI124_AVX2 2 "vector_operand" "xBm,xm")]
UNSPEC_PSIGN))]
"TARGET_SSSE3"
"@
@@ -14399,7 +14400,7 @@
[(set (match_operand:SSESCALARMODE 0 "register_operand" "=x,v")
(unspec:SSESCALARMODE
[(match_operand:SSESCALARMODE 1 "register_operand" "0,v")
- (match_operand:SSESCALARMODE 2 "nonimmediate_operand" "xm,vm")
+ (match_operand:SSESCALARMODE 2 "vector_operand" "xBm,vm")
(match_operand:SI 3 "const_0_to_255_mul_8_operand" "n,n")]
UNSPEC_PALIGNR))]
"TARGET_SSSE3"
@@ -14454,7 +14455,7 @@
(define_insn "*abs<mode>2"
[(set (match_operand:VI1248_AVX512VL_AVX512BW 0 "register_operand" "=v")
(abs:VI1248_AVX512VL_AVX512BW
- (match_operand:VI1248_AVX512VL_AVX512BW 1 "nonimmediate_operand" "vm")))]
+ (match_operand:VI1248_AVX512VL_AVX512BW 1 "vector_operand" "vBm")))]
"TARGET_SSSE3"
"%vpabs<ssemodesuffix>\t{%1, %0|%0, %1}"
[(set_attr "type" "sselog1")
@@ -14492,7 +14493,7 @@
(define_expand "abs<mode>2"
[(set (match_operand:VI1248_AVX512VL_AVX512BW 0 "register_operand")
(abs:VI1248_AVX512VL_AVX512BW
- (match_operand:VI1248_AVX512VL_AVX512BW 1 "nonimmediate_operand")))]
+ (match_operand:VI1248_AVX512VL_AVX512BW 1 "vector_operand")))]
"TARGET_SSE2"
{
if (!TARGET_SSSE3)
@@ -14606,7 +14607,7 @@
(define_insn "<sse4_1>_blend<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF_128_256 0 "register_operand" "=Yr,*x,x")
(vec_merge:VF_128_256
- (match_operand:VF_128_256 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ (match_operand:VF_128_256 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:VF_128_256 1 "register_operand" "0,0,x")
(match_operand:SI 3 "const_0_to_<blendbits>_operand")))]
"TARGET_SSE4_1"
@@ -14626,7 +14627,7 @@
[(set (match_operand:VF_128_256 0 "register_operand" "=Yr,*x,x")
(unspec:VF_128_256
[(match_operand:VF_128_256 1 "register_operand" "0,0,x")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ (match_operand:VF_128_256 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:VF_128_256 3 "register_operand" "Yz,Yz,x")]
UNSPEC_BLENDV))]
"TARGET_SSE4_1"
@@ -14646,8 +14647,8 @@
(define_insn "<sse4_1>_dp<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF_128_256 0 "register_operand" "=Yr,*x,x")
(unspec:VF_128_256
- [(match_operand:VF_128_256 1 "nonimmediate_operand" "%0,0,x")
- (match_operand:VF_128_256 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ [(match_operand:VF_128_256 1 "vector_operand" "%0,0,x")
+ (match_operand:VF_128_256 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:SI 3 "const_0_to_255_operand" "n,n,n")]
UNSPEC_DP))]
"TARGET_SSE4_1"
@@ -14684,7 +14685,7 @@
[(set (match_operand:VI1_AVX2 0 "register_operand" "=Yr,*x,x")
(unspec:VI1_AVX2
[(match_operand:VI1_AVX2 1 "register_operand" "0,0,x")
- (match_operand:VI1_AVX2 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ (match_operand:VI1_AVX2 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:SI 3 "const_0_to_255_operand" "n,n,n")]
UNSPEC_MPSADBW))]
"TARGET_SSE4_1"
@@ -14707,7 +14708,7 @@
(us_truncate:<ssehalfvecmode>
(match_operand:<sseunpackmode> 1 "register_operand" "0,0,v"))
(us_truncate:<ssehalfvecmode>
- (match_operand:<sseunpackmode> 2 "nonimmediate_operand" "Yrm,*xm,vm"))))]
+ (match_operand:<sseunpackmode> 2 "vector_operand" "YrBm,*xBm,vm"))))]
"TARGET_SSE4_1 && <mask_mode512bit_condition> && <mask_avx512bw_condition>"
"@
packusdw\t{%2, %0|%0, %2}
@@ -14723,7 +14724,7 @@
[(set (match_operand:VI1_AVX2 0 "register_operand" "=Yr,*x,x")
(unspec:VI1_AVX2
[(match_operand:VI1_AVX2 1 "register_operand" "0,0,x")
- (match_operand:VI1_AVX2 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ (match_operand:VI1_AVX2 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:VI1_AVX2 3 "register_operand" "Yz,Yz,x")]
UNSPEC_BLENDV))]
"TARGET_SSE4_1"
@@ -14742,7 +14743,7 @@
(define_insn "sse4_1_pblendw"
[(set (match_operand:V8HI 0 "register_operand" "=Yr,*x,x")
(vec_merge:V8HI
- (match_operand:V8HI 2 "nonimmediate_operand" "Yrm,*xm,xm")
+ (match_operand:V8HI 2 "vector_operand" "YrBm,*xBm,xm")
(match_operand:V8HI 1 "register_operand" "0,0,x")
(match_operand:SI 3 "const_0_to_255_operand" "n,n,n")))]
"TARGET_SSE4_1"
@@ -14803,7 +14804,7 @@
(define_insn "sse4_1_phminposuw"
[(set (match_operand:V8HI 0 "register_operand" "=Yr,*x")
- (unspec:V8HI [(match_operand:V8HI 1 "nonimmediate_operand" "Yrm,*xm")]
+ (unspec:V8HI [(match_operand:V8HI 1 "vector_operand" "YrBm,*xBm")]
UNSPEC_PHMINPOSUW))]
"TARGET_SSE4_1"
"%vphminposuw\t{%1, %0|%0, %1}"
@@ -15063,7 +15064,7 @@
(define_insn "<sse4_1>_ptest<mode>"
[(set (reg:CC FLAGS_REG)
(unspec:CC [(match_operand:V_AVX 0 "register_operand" "Yr, *x, x")
- (match_operand:V_AVX 1 "nonimmediate_operand" "Yrm, *xm, xm")]
+ (match_operand:V_AVX 1 "vector_operand" "YrBm, *xBm, xm")]
UNSPEC_PTEST))]
"TARGET_SSE4_1"
"%vptest\t{%1, %0|%0, %1}"
@@ -15081,7 +15082,7 @@
(define_insn "<sse4_1>_round<ssemodesuffix><avxsizesuffix>"
[(set (match_operand:VF_128_256 0 "register_operand" "=Yr,*x")
(unspec:VF_128_256
- [(match_operand:VF_128_256 1 "nonimmediate_operand" "Yrm,*xm")
+ [(match_operand:VF_128_256 1 "vector_operand" "YrBm,*xBm")
(match_operand:SI 2 "const_0_to_15_operand" "n,n")]
UNSPEC_ROUND))]
"TARGET_ROUND"
@@ -15099,7 +15100,7 @@
(define_expand "<sse4_1>_round<ssemodesuffix>_sfix<avxsizesuffix>"
[(match_operand:<sseintvecmode> 0 "register_operand")
- (match_operand:VF1_128_256 1 "nonimmediate_operand")
+ (match_operand:VF1_128_256 1 "vector_operand")
(match_operand:SI 2 "const_0_to_15_operand")]
"TARGET_ROUND"
{
@@ -15125,8 +15126,8 @@
(define_expand "<sse4_1>_round<ssemodesuffix>_vec_pack_sfix<avxsizesuffix>"
[(match_operand:<ssepackfltmode> 0 "register_operand")
- (match_operand:VF2 1 "nonimmediate_operand")
- (match_operand:VF2 2 "nonimmediate_operand")
+ (match_operand:VF2 1 "vector_operand")
+ (match_operand:VF2 2 "vector_operand")
(match_operand:SI 3 "const_0_to_15_operand")]
"TARGET_ROUND"
{
@@ -16926,7 +16927,7 @@
(define_insn "aesenc"
[(set (match_operand:V2DI 0 "register_operand" "=x,x")
(unspec:V2DI [(match_operand:V2DI 1 "register_operand" "0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,xm")]
+ (match_operand:V2DI 2 "vector_operand" "xBm,xm")]
UNSPEC_AESENC))]
"TARGET_AES"
"@
@@ -16942,7 +16943,7 @@
(define_insn "aesenclast"
[(set (match_operand:V2DI 0 "register_operand" "=x,x")
(unspec:V2DI [(match_operand:V2DI 1 "register_operand" "0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,xm")]
+ (match_operand:V2DI 2 "vector_operand" "xBm,xm")]
UNSPEC_AESENCLAST))]
"TARGET_AES"
"@
@@ -16958,7 +16959,7 @@
(define_insn "aesdec"
[(set (match_operand:V2DI 0 "register_operand" "=x,x")
(unspec:V2DI [(match_operand:V2DI 1 "register_operand" "0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,xm")]
+ (match_operand:V2DI 2 "vector_operand" "xBm,xm")]
UNSPEC_AESDEC))]
"TARGET_AES"
"@
@@ -16974,7 +16975,7 @@
(define_insn "aesdeclast"
[(set (match_operand:V2DI 0 "register_operand" "=x,x")
(unspec:V2DI [(match_operand:V2DI 1 "register_operand" "0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,xm")]
+ (match_operand:V2DI 2 "vector_operand" "xBm,xm")]
UNSPEC_AESDECLAST))]
"TARGET_AES"
"@
@@ -16989,7 +16990,7 @@
(define_insn "aesimc"
[(set (match_operand:V2DI 0 "register_operand" "=x")
- (unspec:V2DI [(match_operand:V2DI 1 "nonimmediate_operand" "xm")]
+ (unspec:V2DI [(match_operand:V2DI 1 "vector_operand" "xBm")]
UNSPEC_AESIMC))]
"TARGET_AES"
"%vaesimc\t{%1, %0|%0, %1}"
@@ -17000,7 +17001,7 @@
(define_insn "aeskeygenassist"
[(set (match_operand:V2DI 0 "register_operand" "=x")
- (unspec:V2DI [(match_operand:V2DI 1 "nonimmediate_operand" "xm")
+ (unspec:V2DI [(match_operand:V2DI 1 "vector_operand" "xBm")
(match_operand:SI 2 "const_0_to_255_operand" "n")]
UNSPEC_AESKEYGENASSIST))]
"TARGET_AES"
@@ -17014,7 +17015,7 @@
(define_insn "pclmulqdq"
[(set (match_operand:V2DI 0 "register_operand" "=x,x")
(unspec:V2DI [(match_operand:V2DI 1 "register_operand" "0,x")
- (match_operand:V2DI 2 "nonimmediate_operand" "xm,xm")
+ (match_operand:V2DI 2 "vector_operand" "xBm,xm")
(match_operand:SI 3 "const_0_to_255_operand" "n,n")]
UNSPEC_PCLMUL))]
"TARGET_PCLMUL"
@@ -19152,7 +19153,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")]
+ (match_operand:V4SI 2 "vector_operand" "xBm")]
UNSPEC_SHA1MSG1))]
"TARGET_SHA"
"sha1msg1\t{%2, %0|%0, %2}"
@@ -19163,7 +19164,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")]
+ (match_operand:V4SI 2 "vector_operand" "xBm")]
UNSPEC_SHA1MSG2))]
"TARGET_SHA"
"sha1msg2\t{%2, %0|%0, %2}"
@@ -19174,7 +19175,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")]
+ (match_operand:V4SI 2 "vector_operand" "xBm")]
UNSPEC_SHA1NEXTE))]
"TARGET_SHA"
"sha1nexte\t{%2, %0|%0, %2}"
@@ -19185,7 +19186,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")
+ (match_operand:V4SI 2 "vector_operand" "xBm")
(match_operand:SI 3 "const_0_to_3_operand" "n")]
UNSPEC_SHA1RNDS4))]
"TARGET_SHA"
@@ -19198,7 +19199,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")]
+ (match_operand:V4SI 2 "vector_operand" "xBm")]
UNSPEC_SHA256MSG1))]
"TARGET_SHA"
"sha256msg1\t{%2, %0|%0, %2}"
@@ -19209,7 +19210,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")]
+ (match_operand:V4SI 2 "vector_operand" "xBm")]
UNSPEC_SHA256MSG2))]
"TARGET_SHA"
"sha256msg2\t{%2, %0|%0, %2}"
@@ -19220,7 +19221,7 @@
[(set (match_operand:V4SI 0 "register_operand" "=x")
(unspec:V4SI
[(match_operand:V4SI 1 "register_operand" "0")
- (match_operand:V4SI 2 "nonimmediate_operand" "xm")
+ (match_operand:V4SI 2 "vector_operand" "xBm")
(match_operand:V4SI 3 "register_operand" "Yz")]
UNSPEC_SHA256RNDS2))]
"TARGET_SHA"
diff --git a/gcc/config/i386/subst.md b/gcc/config/i386/subst.md
index b283dd8..d35f34c 100644
--- a/gcc/config/i386/subst.md
+++ b/gcc/config/i386/subst.md
@@ -123,7 +123,7 @@
(define_subst_attr "round_constraint" "round" "vm" "v")
(define_subst_attr "round_constraint2" "round" "m" "v")
(define_subst_attr "round_constraint3" "round" "rm" "r")
-(define_subst_attr "round_nimm_predicate" "round" "nonimmediate_operand" "register_operand")
+(define_subst_attr "round_nimm_predicate" "round" "vector_operand" "register_operand")
(define_subst_attr "round_prefix" "round" "vex" "evex")
(define_subst_attr "round_mode512bit_condition" "round" "1" "(<MODE>mode == V16SFmode
|| <MODE>mode == V8DFmode
@@ -162,7 +162,8 @@
(define_subst_attr "round_saeonly_sd_mask_op5" "round_saeonly" "" "<round_saeonly_sd_mask_operand5>")
(define_subst_attr "round_saeonly_constraint" "round_saeonly" "vm" "v")
(define_subst_attr "round_saeonly_constraint2" "round_saeonly" "m" "v")
-(define_subst_attr "round_saeonly_nimm_predicate" "round_saeonly" "nonimmediate_operand" "register_operand")
+(define_subst_attr "round_saeonly_nimm_predicate" "round_saeonly" "vector_operand" "register_operand")
+(define_subst_attr "round_saeonly_nimm_scalar_predicate" "round_saeonly" "nonimmediate_operand" "register_operand")
(define_subst_attr "round_saeonly_mode512bit_condition" "round_saeonly" "1" "(<MODE>mode == V16SFmode
|| <MODE>mode == V8DFmode
|| <MODE>mode == V8DImode
diff --git a/gcc/testsuite/gcc.target/i386/pr68991.c b/gcc/testsuite/gcc.target/i386/pr68991.c
new file mode 100644
index 0000000..3fc5ac8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr68991.c
@@ -0,0 +1,32 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -msse2 -mfpmath=sse" } */
+
+float
+foo (float a, float b, float c, float d)
+{
+ float ac, bd, ad, bc, y;
+
+ ac = a * c;
+ bd = b * d;
+ ad = a * d;
+ bc = b * c;
+
+ if (__builtin_expect (!__builtin_expect ((a) != (a), 0)
+ & !__builtin_expect (!__builtin_expect (((a) - (a)) != ((a) - (a)), 0), 1), 0)
+ || __builtin_expect (!__builtin_expect ((b) != (b), 0)
+ & !__builtin_expect (!__builtin_expect (((b) - (b)) != ((b) - (b)), 0), 1), 0))
+ a = __builtin_copysignf (__builtin_expect (!__builtin_expect ((a) != (a), 0)
+ & !__builtin_expect (!__builtin_expect (((a) - (a)) != ((a) - (a)), 0), 1), 0) ? 1 : 0, a);
+
+ c = __builtin_copysignf (__builtin_expect (!__builtin_expect ((c) != (c), 0) & !__builtin_expect (!__builtin_expect (((c) - (c)) != ((c) - (c)), 0), 1), 0) ? 1 : 0, c);
+ if ((__builtin_expect (!__builtin_expect ((ac) != (ac), 0)
+ & !__builtin_expect (!__builtin_expect (((ac) - (ac)) != ((ac) - (ac)), 0), 1), 0)
+ || __builtin_expect (!__builtin_expect ((bd) != (bd), 0)
+ & !__builtin_expect (!__builtin_expect (((bd) - (bd)) != ((bd) - (bd)), 0), 1), 0)
+ || __builtin_expect (!__builtin_expect ((bc) != (bc), 0) & !__builtin_expect (!__builtin_expect (((bc) - (bc)) != ((bc) - (bc)), 0), 1), 0)))
+ d = __builtin_copysignf (0, d);
+
+ y = a * d + b * c;
+
+ return y;
+}
--
2.5.0