Implement vldr?q_gather_base using the new MVE builtins framework.
The patch updates two testcases rather than using different iterators
for predicated and non-predicated versions. According to ACLE:
vldrdq_gather_base_s64 is expected to generate VLDRD.64
vldrdq_gather_base_z_s64 is expected to generate VLDRDT.U64
Both are equally valid, however.
gcc/ChangeLog:
* config/arm/arm-builtins.cc (arm_ldrgbs_qualifiers)
(arm_ldrgbu_qualifiers, arm_ldrgbs_z_qualifiers)
(arm_ldrgbu_z_qualifiers): Delete.
* config/arm/arm-mve-builtins-base.cc (class
vldrq_gather_base_impl): New.
(vldrdq_gather_base, vldrwq_gather_base): New.
* config/arm/arm-mve-builtins-base.def (vldrdq_gather_base)
(vldrwq_gather_base): New.
* config/arm/arm-mve-builtins-base.h: (vldrdq_gather_base)
(vldrwq_gather_base): New.
* config/arm/arm_mve.h (vldrwq_gather_base_s32): Delete.
(vldrwq_gather_base_u32): Delete.
(vldrwq_gather_base_z_u32): Delete.
(vldrwq_gather_base_z_s32): Delete.
(vldrdq_gather_base_s64): Delete.
(vldrdq_gather_base_u64): Delete.
(vldrdq_gather_base_z_s64): Delete.
(vldrdq_gather_base_z_u64): Delete.
(vldrwq_gather_base_f32): Delete.
(vldrwq_gather_base_z_f32): Delete.
(__arm_vldrwq_gather_base_s32): Delete.
(__arm_vldrwq_gather_base_u32): Delete.
(__arm_vldrwq_gather_base_z_s32): Delete.
(__arm_vldrwq_gather_base_z_u32): Delete.
(__arm_vldrdq_gather_base_s64): Delete.
(__arm_vldrdq_gather_base_u64): Delete.
(__arm_vldrdq_gather_base_z_s64): Delete.
(__arm_vldrdq_gather_base_z_u64): Delete.
(__arm_vldrwq_gather_base_f32): Delete.
(__arm_vldrwq_gather_base_z_f32): Delete.
* config/arm/arm_mve_builtins.def (vldrwq_gather_base_s)
(vldrwq_gather_base_u, vldrwq_gather_base_z_s)
(vldrwq_gather_base_z_u, vldrdq_gather_base_s)
(vldrwq_gather_base_f, vldrdq_gather_base_z_s)
(vldrwq_gather_base_z_f, vldrdq_gather_base_u)
(vldrdq_gather_base_z_u): Delete.
* config/arm/iterators.md (supf): Remove VLDRWQGB_S, VLDRWQGB_U,
VLDRDQGB_S, VLDRDQGB_U.
(VLDRWGBQ, VLDRDGBQ): Delete.
* config/arm/mve.md (mve_vldrwq_gather_base_<supf>v4si): Delete.
(mve_vldrwq_gather_base_z_<supf>v4si): Delete.
(mve_vldrdq_gather_base_<supf>v2di): Delete.
(mve_vldrdq_gather_base_z_<supf>v2di): Delete.
(mve_vldrwq_gather_base_fv4sf): Delete.
(mve_vldrwq_gather_base_z_fv4sf): Delete.
(@mve_vldrq_gather_base_<mode>): New.
(@mve_vldrq_gather_base_z_<mode>): New.
* config/arm/unspecs.md (VLDRWQGB_S, VLDRWQGB_U, VLDRDQGB_S)
(VLDRDQGB_U, VLDRWQGB_F): Delete.
(VLDRGBQ, VLDRGBQ_Z): New.
gcc/testsuite/ChangeLog:
* gcc.target/arm/mve/intrinsics/vldrdq_gather_base_s64.c: Update
expected output.
* gcc.target/arm/mve/intrinsics/vldrdq_gather_base_u64.c:
Likewise.
#define QUADOP_UNONE_UNONE_UNONE_NONE_PRED_QUALIFIERS \
(arm_quadop_unone_unone_unone_none_pred_qualifiers)
-static enum arm_type_qualifiers
-arm_ldrgbs_qualifiers[SIMD_MAX_BUILTIN_ARGS]
- = { qualifier_none, qualifier_unsigned, qualifier_immediate};
-#define LDRGBS_QUALIFIERS (arm_ldrgbs_qualifiers)
-
-static enum arm_type_qualifiers
-arm_ldrgbu_qualifiers[SIMD_MAX_BUILTIN_ARGS]
- = { qualifier_unsigned, qualifier_unsigned, qualifier_immediate};
-#define LDRGBU_QUALIFIERS (arm_ldrgbu_qualifiers)
-
-static enum arm_type_qualifiers
-arm_ldrgbs_z_qualifiers[SIMD_MAX_BUILTIN_ARGS]
- = { qualifier_none, qualifier_unsigned, qualifier_immediate,
- qualifier_predicate};
-#define LDRGBS_Z_QUALIFIERS (arm_ldrgbs_z_qualifiers)
-
-static enum arm_type_qualifiers
-arm_ldrgbu_z_qualifiers[SIMD_MAX_BUILTIN_ARGS]
- = { qualifier_unsigned, qualifier_unsigned, qualifier_immediate,
- qualifier_predicate};
-#define LDRGBU_Z_QUALIFIERS (arm_ldrgbu_z_qualifiers)
-
static enum arm_type_qualifiers
arm_ldrgbwbxu_qualifiers[SIMD_MAX_BUILTIN_ARGS]
= { qualifier_unsigned, qualifier_unsigned, qualifier_immediate};
}
};
+ /* Builds the vldrq_gather_base intrinsics. */
+class vldrq_gather_base_impl : public load_extending
+{
+public:
+ using load_extending::load_extending;
+
+ rtx expand (function_expander &e) const override
+ {
+ insn_code icode;
+ rtx insns;
+
+ switch (e.pred)
+ {
+ case PRED_none:
+ icode = code_for_mve_vldrq_gather_base (e.vector_mode (0));
+ break;
+
+ case PRED_z:
+ icode = code_for_mve_vldrq_gather_base_z (e.vector_mode (0));
+ break;
+
+ default:
+ gcc_unreachable ();
+ }
+ insns = e.use_exact_insn (icode);
+
+ return insns;
+ }
+};
+
/* Implements vctp8q, vctp16q, vctp32q and vctp64q intrinsics. */
class vctpq_impl : public function_base
{
FUNCTION (vldrbq, vldrq_impl, (TYPE_SUFFIX_s8, TYPE_SUFFIX_u8))
FUNCTION (vldrbq_gather, vldrq_gather_impl, (false, TYPE_SUFFIX_s8, TYPE_SUFFIX_u8))
FUNCTION (vldrdq_gather, vldrq_gather_impl, (false, TYPE_SUFFIX_s64, TYPE_SUFFIX_u64, NUM_TYPE_SUFFIXES))
+FUNCTION (vldrdq_gather_base, vldrq_gather_base_impl, (TYPE_SUFFIX_s64, TYPE_SUFFIX_u64))
FUNCTION (vldrdq_gather_shifted, vldrq_gather_impl, (true, TYPE_SUFFIX_s64, TYPE_SUFFIX_u64, NUM_TYPE_SUFFIXES))
FUNCTION (vldrhq, vldrq_impl, (TYPE_SUFFIX_s16, TYPE_SUFFIX_u16, TYPE_SUFFIX_f16))
FUNCTION (vldrhq_gather, vldrq_gather_impl, (false, TYPE_SUFFIX_s16, TYPE_SUFFIX_u16, TYPE_SUFFIX_f16))
FUNCTION (vldrhq_gather_shifted, vldrq_gather_impl, (true, TYPE_SUFFIX_s16, TYPE_SUFFIX_u16, TYPE_SUFFIX_f16))
FUNCTION (vldrwq, vldrq_impl, (TYPE_SUFFIX_s32, TYPE_SUFFIX_u32, TYPE_SUFFIX_f32))
FUNCTION (vldrwq_gather, vldrq_gather_impl, (false, TYPE_SUFFIX_s32, TYPE_SUFFIX_u32, TYPE_SUFFIX_f32))
+FUNCTION (vldrwq_gather_base, vldrq_gather_base_impl, (TYPE_SUFFIX_s32, TYPE_SUFFIX_u32, TYPE_SUFFIX_f32))
FUNCTION (vldrwq_gather_shifted, vldrq_gather_impl, (true, TYPE_SUFFIX_s32, TYPE_SUFFIX_u32, TYPE_SUFFIX_f32))
FUNCTION_PRED_P_S (vmaxavq, VMAXAVQ)
FUNCTION_WITHOUT_N_NO_U_F (vmaxaq, VMAXAQ)
DEF_MVE_FUNCTION (vldrbq, load_ext, all_integer, z_or_none)
DEF_MVE_FUNCTION (vldrbq_gather, load_ext_gather_offset, all_integer, z_or_none)
DEF_MVE_FUNCTION (vldrdq_gather, load_ext_gather_offset, integer_64, z_or_none)
+DEF_MVE_FUNCTION (vldrdq_gather_base, load_gather_base, integer_64, z_or_none)
DEF_MVE_FUNCTION (vldrdq_gather_shifted, load_ext_gather_offset, integer_64, z_or_none)
DEF_MVE_FUNCTION (vldrhq, load_ext, integer_16_32, z_or_none)
DEF_MVE_FUNCTION (vldrhq_gather, load_ext_gather_offset, integer_16_32, z_or_none)
DEF_MVE_FUNCTION (vldrhq_gather_shifted, load_ext_gather_offset, integer_16_32, z_or_none)
DEF_MVE_FUNCTION (vldrwq, load_ext, integer_32, z_or_none)
DEF_MVE_FUNCTION (vldrwq_gather, load_ext_gather_offset, integer_32, z_or_none)
+DEF_MVE_FUNCTION (vldrwq_gather_base, load_gather_base, integer_32, z_or_none)
DEF_MVE_FUNCTION (vldrwq_gather_shifted, load_ext_gather_offset, integer_32, z_or_none)
DEF_MVE_FUNCTION (vmaxaq, binary_maxamina, all_signed, m_or_none)
DEF_MVE_FUNCTION (vmaxavq, binary_maxavminav, all_signed, p_or_none)
DEF_MVE_FUNCTION (vldrhq_gather_shifted, load_ext_gather_offset, float_16, z_or_none)
DEF_MVE_FUNCTION (vldrwq, load_ext, float_32, z_or_none)
DEF_MVE_FUNCTION (vldrwq_gather, load_ext_gather_offset, float_32, z_or_none)
+DEF_MVE_FUNCTION (vldrwq_gather_base, load_gather_base, float_32, z_or_none)
DEF_MVE_FUNCTION (vldrwq_gather_shifted, load_ext_gather_offset, float_32, z_or_none)
DEF_MVE_FUNCTION (vmaxnmaq, binary, all_float, m_or_none)
DEF_MVE_FUNCTION (vmaxnmavq, binary_maxvminv, all_float, p_or_none)
extern const function_base *const vldrbq;
extern const function_base *const vldrbq_gather;
extern const function_base *const vldrdq_gather;
+extern const function_base *const vldrdq_gather_base;
extern const function_base *const vldrdq_gather_shifted;
extern const function_base *const vldrhq;
extern const function_base *const vldrhq_gather;
extern const function_base *const vldrhq_gather_shifted;
extern const function_base *const vldrwq;
extern const function_base *const vldrwq_gather;
+extern const function_base *const vldrwq_gather_base;
extern const function_base *const vldrwq_gather_shifted;
extern const function_base *const vmaxaq;
extern const function_base *const vmaxavq;
#define vst4q_f16( __addr, __value) __arm_vst4q_f16( __addr, __value)
#define vst4q_f32( __addr, __value) __arm_vst4q_f32( __addr, __value)
#define vpnot(__a) __arm_vpnot(__a)
-#define vldrwq_gather_base_s32(__addr, __offset) __arm_vldrwq_gather_base_s32(__addr, __offset)
-#define vldrwq_gather_base_u32(__addr, __offset) __arm_vldrwq_gather_base_u32(__addr, __offset)
-#define vldrwq_gather_base_z_u32(__addr, __offset, __p) __arm_vldrwq_gather_base_z_u32(__addr, __offset, __p)
-#define vldrwq_gather_base_z_s32(__addr, __offset, __p) __arm_vldrwq_gather_base_z_s32(__addr, __offset, __p)
-#define vldrdq_gather_base_s64(__addr, __offset) __arm_vldrdq_gather_base_s64(__addr, __offset)
-#define vldrdq_gather_base_u64(__addr, __offset) __arm_vldrdq_gather_base_u64(__addr, __offset)
-#define vldrdq_gather_base_z_s64(__addr, __offset, __p) __arm_vldrdq_gather_base_z_s64(__addr, __offset, __p)
-#define vldrdq_gather_base_z_u64(__addr, __offset, __p) __arm_vldrdq_gather_base_z_u64(__addr, __offset, __p)
-#define vldrwq_gather_base_f32(__addr, __offset) __arm_vldrwq_gather_base_f32(__addr, __offset)
-#define vldrwq_gather_base_z_f32(__addr, __offset, __p) __arm_vldrwq_gather_base_z_f32(__addr, __offset, __p)
#define vuninitializedq_u8(void) __arm_vuninitializedq_u8(void)
#define vuninitializedq_u16(void) __arm_vuninitializedq_u16(void)
#define vuninitializedq_u32(void) __arm_vuninitializedq_u32(void)
return __builtin_mve_vpnotv16bi (__a);
}
-__extension__ extern __inline int32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_s32 (uint32x4_t __addr, const int __offset)
-{
- return __builtin_mve_vldrwq_gather_base_sv4si (__addr, __offset);
-}
-
-__extension__ extern __inline uint32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_u32 (uint32x4_t __addr, const int __offset)
-{
- return __builtin_mve_vldrwq_gather_base_uv4si (__addr, __offset);
-}
-
-__extension__ extern __inline int32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_z_s32 (uint32x4_t __addr, const int __offset, mve_pred16_t __p)
-{
- return __builtin_mve_vldrwq_gather_base_z_sv4si (__addr, __offset, __p);
-}
-
-__extension__ extern __inline uint32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_z_u32 (uint32x4_t __addr, const int __offset, mve_pred16_t __p)
-{
- return __builtin_mve_vldrwq_gather_base_z_uv4si (__addr, __offset, __p);
-}
-
-__extension__ extern __inline int64x2_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrdq_gather_base_s64 (uint64x2_t __addr, const int __offset)
-{
- return __builtin_mve_vldrdq_gather_base_sv2di (__addr, __offset);
-}
-
-__extension__ extern __inline uint64x2_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrdq_gather_base_u64 (uint64x2_t __addr, const int __offset)
-{
- return __builtin_mve_vldrdq_gather_base_uv2di (__addr, __offset);
-}
-
-__extension__ extern __inline int64x2_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrdq_gather_base_z_s64 (uint64x2_t __addr, const int __offset, mve_pred16_t __p)
-{
- return __builtin_mve_vldrdq_gather_base_z_sv2di (__addr, __offset, __p);
-}
-
-__extension__ extern __inline uint64x2_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrdq_gather_base_z_u64 (uint64x2_t __addr, const int __offset, mve_pred16_t __p)
-{
- return __builtin_mve_vldrdq_gather_base_z_uv2di (__addr, __offset, __p);
-}
-
__extension__ extern __inline int64x2_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vldrdq_gather_base_wb_s64 (uint64x2_t * __addr, const int __offset)
__builtin_mve_vst4qv4sf (__addr, __rv.__o);
}
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_f32 (uint32x4_t __addr, const int __offset)
-{
- return __builtin_mve_vldrwq_gather_base_fv4sf (__addr, __offset);
-}
-
-__extension__ extern __inline float32x4_t
-__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-__arm_vldrwq_gather_base_z_f32 (uint32x4_t __addr, const int __offset, mve_pred16_t __p)
-{
- return __builtin_mve_vldrwq_gather_base_z_fv4sf (__addr, __offset, __p);
-}
-
__extension__ extern __inline float32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
__arm_vldrwq_gather_base_wb_f32 (uint32x4_t * __addr, const int __offset)
VAR2 (QUADOP_NONE_NONE_NONE_NONE_PRED, vaddq_m_n_f, v8hf, v4sf)
VAR2 (QUADOP_NONE_NONE_NONE_NONE_PRED, vaddq_m_f, v8hf, v4sf)
VAR2 (QUADOP_NONE_NONE_NONE_NONE_PRED, vabdq_m_f, v8hf, v4sf)
-VAR1 (LDRGBS, vldrwq_gather_base_s, v4si)
-VAR1 (LDRGBU, vldrwq_gather_base_u, v4si)
-VAR1 (LDRGBS_Z, vldrwq_gather_base_z_s, v4si)
-VAR1 (LDRGBU_Z, vldrwq_gather_base_z_u, v4si)
-VAR1 (LDRGBS, vldrdq_gather_base_s, v2di)
-VAR1 (LDRGBS, vldrwq_gather_base_f, v4sf)
-VAR1 (LDRGBS_Z, vldrdq_gather_base_z_s, v2di)
-VAR1 (LDRGBS_Z, vldrwq_gather_base_z_f, v4sf)
-VAR1 (LDRGBU, vldrdq_gather_base_u, v2di)
-VAR1 (LDRGBU_Z, vldrdq_gather_base_z_u, v2di)
VAR1 (LDRGBWBU_Z, vldrwq_gather_base_nowb_z_u, v4si)
VAR1 (LDRGBWBU_Z, vldrdq_gather_base_nowb_z_u, v2di)
VAR1 (LDRGBWBU, vldrwq_gather_base_nowb_u, v4si)
(VQRSHRNBQ_M_N_S "s") (VQRSHRNBQ_M_N_U "u")
(VMLALDAVAXQ_P_S "s")
(VMLALDAVAQ_P_S "s") (VMLALDAVAQ_P_U "u")
- (VLDRWQGB_S "s")
- (VLDRWQGB_U "u")
- (VLDRDQGB_S "s") (VLDRDQGB_U "u")
(VSTRDQSB_S "s") (VSTRDQSB_U "u")
(VLDRWQGBWB_S "s") (VLDRWQGBWB_U "u") (VLDRDQGBWB_S "s")
(VLDRDQGBWB_U "u") (VADCQ_M_S "s")
(define_int_iterator VSHLLxQ_M_N [VSHLLBQ_M_N_U VSHLLBQ_M_N_S VSHLLTQ_M_N_U VSHLLTQ_M_N_S])
(define_int_iterator VSHRNBQ_M_N [VSHRNBQ_M_N_S VSHRNBQ_M_N_U])
(define_int_iterator VSHRNTQ_M_N [VSHRNTQ_M_N_S VSHRNTQ_M_N_U])
-(define_int_iterator VLDRWGBQ [VLDRWQGB_S VLDRWQGB_U])
-(define_int_iterator VLDRDGBQ [VLDRDQGB_S VLDRDQGB_U])
(define_int_iterator VLDRWGBWBQ [VLDRWQGBWB_S VLDRWQGBWB_U])
(define_int_iterator VLDRDGBWBQ [VLDRDQGBWB_S VLDRDQGBWB_U])
(define_int_iterator VxCIQ [VADCIQ_U VADCIQ_S VSBCIQ_U VSBCIQ_S])
[(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrq_gather_offset_extend_<mode><US>"))
(set_attr "length" "8")])
-;;
-;; [vldrwq_gather_base_s vldrwq_gather_base_u]
-;;
-(define_insn "mve_vldrwq_gather_base_<supf>v4si"
- [(set (match_operand:V4SI 0 "s_register_operand" "=&w")
- (unspec:V4SI [(match_operand:V4SI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")]
- VLDRWGBQ))
- ]
- "TARGET_HAVE_MVE"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vldrw.u32\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrwq_gather_base_<supf>v4si"))
- (set_attr "length" "4")])
-
;; Predicated vector scatter stores with base
;;
;; [vstrdq_scatter_base_p_s vstrdq_scatter_base_p_u]
[(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vstrq_scatter_base_<mode>"))
(set_attr "length" "8")])
+;; Vector gather loads with base
+;;
+;; [vldrwq_gather_base_s vldrwq_gather_base_u]
+;; [vldrwq_gather_base_f]
+;; [vldrdq_gather_base_s vldrdq_gather_base_u]
+;;
+(define_insn "@mve_vldrq_gather_base_<mode>"
+ [(set (match_operand:MVE_4 0 "s_register_operand" "=&w")
+ (unspec:MVE_4 [(match_operand:<MVE_scatter_offset> 1 "s_register_operand" "w")
+ (match_operand:SI 2 "immediate_operand" "i")]
+ VLDRGBQ))
+ ]
+ "(TARGET_HAVE_MVE && VALID_MVE_SI_MODE (<MODE>mode))
+ || (TARGET_HAVE_MVE_FLOAT && VALID_MVE_SF_MODE (<MODE>mode))"
+ "vldr<MVE_elem_ch>.u<V_sz_elem>\t%q0, [%q1, %2]"
+ [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrq_gather_base_<mode>"))
+ (set_attr "length" "4")])
+
+;; Predicated vector gather loads with base
;;
;; [vldrwq_gather_base_z_s vldrwq_gather_base_z_u]
+;; [vldrwq_gather_base_z_f]
+;; [vldrdq_gather_base_z_s vldrdq_gather_base_z_u]
;;
-(define_insn "mve_vldrwq_gather_base_z_<supf>v4si"
- [(set (match_operand:V4SI 0 "s_register_operand" "=&w")
- (unspec:V4SI [(match_operand:V4SI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")
- (match_operand:V4BI 3 "vpr_register_operand" "Up")]
- VLDRWGBQ))
+(define_insn "@mve_vldrq_gather_base_z_<mode>"
+ [(set (match_operand:MVE_4 0 "s_register_operand" "=&w")
+ (unspec:MVE_4 [(match_operand:<MVE_scatter_offset> 1 "s_register_operand" "w")
+ (match_operand:SI 2 "immediate_operand" "i")
+ (match_operand:<MVE_VPRED> 3 "vpr_register_operand" "Up")]
+ VLDRGBQ_Z))
]
- "TARGET_HAVE_MVE"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vpst\n\tvldrwt.u32\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrwq_gather_base_<supf>v4si"))
+ "(TARGET_HAVE_MVE && VALID_MVE_SI_MODE (<MODE>mode))
+ || (TARGET_HAVE_MVE_FLOAT && VALID_MVE_SF_MODE (<MODE>mode))"
+ "vpst\n\tvldr<MVE_elem_ch>t.u<V_sz_elem>\t%q0, [%q1, %2]"
+ [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrq_gather_base_<mode>"))
(set_attr "length" "8")])
+
;; Gather loads with shifted offset
;;
;; [vldrhq_gather_shifted_offset_s vldrhq_gather_shifted_offset_u]
[(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrq_gather_shifted_offset_extend_v4si<US>"))
(set_attr "length" "4")])
-;;
-;; [vldrdq_gather_base_s vldrdq_gather_base_u]
-;;
-(define_insn "mve_vldrdq_gather_base_<supf>v2di"
- [(set (match_operand:V2DI 0 "s_register_operand" "=&w")
- (unspec:V2DI [(match_operand:V2DI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")]
- VLDRDGBQ))
- ]
- "TARGET_HAVE_MVE"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vldrd.64\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrdq_gather_base_<supf>v2di"))
- (set_attr "length" "4")])
-
-;;
-;; [vldrdq_gather_base_z_s vldrdq_gather_base_z_u]
-;;
-(define_insn "mve_vldrdq_gather_base_z_<supf>v2di"
- [(set (match_operand:V2DI 0 "s_register_operand" "=&w")
- (unspec:V2DI [(match_operand:V2DI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")
- (match_operand:V2QI 3 "vpr_register_operand" "Up")]
- VLDRDGBQ))
- ]
- "TARGET_HAVE_MVE"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vpst\n\tvldrdt.u64\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrdq_gather_base_<supf>v2di"))
- (set_attr "length" "8")])
-
-;;
-;; [vldrwq_gather_base_f]
-;;
-(define_insn "mve_vldrwq_gather_base_fv4sf"
- [(set (match_operand:V4SF 0 "s_register_operand" "=&w")
- (unspec:V4SF [(match_operand:V4SI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")]
- VLDRWQGB_F))
- ]
- "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vldrw.u32\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrwq_gather_base_fv4sf"))
- (set_attr "length" "4")])
-
-;;
-;; [vldrwq_gather_base_z_f]
-;;
-(define_insn "mve_vldrwq_gather_base_z_fv4sf"
- [(set (match_operand:V4SF 0 "s_register_operand" "=&w")
- (unspec:V4SF [(match_operand:V4SI 1 "s_register_operand" "w")
- (match_operand:SI 2 "immediate_operand" "i")
- (match_operand:V4BI 3 "vpr_register_operand" "Up")]
- VLDRWQGB_F))
- ]
- "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
-{
- rtx ops[3];
- ops[0] = operands[0];
- ops[1] = operands[1];
- ops[2] = operands[2];
- output_asm_insn ("vpst\n\tvldrwt.u32\t%q0, [%q1, %2]",ops);
- return "";
-}
- [(set (attr "mve_unpredicated_insn") (symbol_ref "CODE_FOR_mve_vldrwq_gather_base_fv4sf"))
- (set_attr "length" "8")])
-
;; Vector scatter stores with shifted offset
;;
;; [vstrhq_scatter_shifted_offset_s vstrhq_scatter_shifted_offset_u]
VLDRGOQ_Z
VLDRGOQ_EXT
VLDRGOQ_EXT_Z
- VLDRWQGB_S
- VLDRWQGB_U
+ VLDRGBQ
+ VLDRGBQ_Z
VLDRGSOQ
VLDRGSOQ_Z
VLDRGSOQ_EXT
VLDRGSOQ_EXT_Z
- VLDRDQGB_S
- VLDRDQGB_U
- VLDRWQGB_F
VSTRQ
VSTRQ_P
VSTRQ_TRUNC
/*
**foo:
** ...
-** vldrd.64 q[0-9]+, \[q[0-9]+, #[0-9]+\](?: @.*|)
+** vldrd.u64 q[0-9]+, \[q[0-9]+, #[0-9]+\](?: @.*|)
** ...
*/
int64x2_t
}
#endif
-/* { dg-final { scan-assembler-not "__ARM_undef" } } */
\ No newline at end of file
+/* { dg-final { scan-assembler-not "__ARM_undef" } } */
/*
**foo:
** ...
-** vldrd.64 q[0-9]+, \[q[0-9]+, #[0-9]+\](?: @.*|)
+** vldrd.u64 q[0-9]+, \[q[0-9]+, #[0-9]+\](?: @.*|)
** ...
*/
uint64x2_t
}
#endif
-/* { dg-final { scan-assembler-not "__ARM_undef" } } */
\ No newline at end of file
+/* { dg-final { scan-assembler-not "__ARM_undef" } } */
projects / thirdparty / gcc.git / commitdiff
