FWIW, the series is on forgejo too:
https://forge.sourceware.org/gcc/gcc-TEST/pulls/68
On Wed, 27 Aug 2025 at 16:45, Christophe Lyon
<christophe.l...@linaro.org> wrote:
>
> Implement vgetq_lane and vsetq_lane using the new MVE builtins
> framework.
>
> Although MVE intrinsics are not supported in big-endian mode, we keep
> the code to convert lane indices into GCC's vector indices, so that
> it's already in place in case we want to support big-endian in the
> future.
>
> The patch adds new tests, to check that we emit an error when the lane
> number is not in the expected range.
>
> gcc/ChangeLog:
>
> * config/arm/arm-mve-builtins-base.cc (class
> mve_function_vset_vget_lane): New.
> (vgetq_lane, vsetq_lane): New.
> * config/arm/arm-mve-builtins-base.def (vgetq_lane, vsetq_lane):
> New.
> * config/arm/arm-mve-builtins-base.h (vgetq_lane, vsetq_lane):
> New.
>
> * config/arm/arm-mve-builtins-shapes.cc (struct getq_lane)
> (setq_lane): New.
> * config/arm/arm-mve-builtins-shapes.h (getq_lane, setq_lane):
> New.
> * config/arm/arm_mve.h (vsetq_lane): Delete.
> (vgetq_lane): Delete.
> (vsetq_lane_f16): Delete.
> (vsetq_lane_f32): Delete.
> (vsetq_lane_s16): Delete.
> (vsetq_lane_s32): Delete.
> (vsetq_lane_s8): Delete.
> (vsetq_lane_s64): Delete.
> (vsetq_lane_u8): Delete.
> (vsetq_lane_u16): Delete.
> (vsetq_lane_u32): Delete.
> (vsetq_lane_u64): Delete.
> (vgetq_lane_f16): Delete.
> (vgetq_lane_f32): Delete.
> (vgetq_lane_s16): Delete.
> (vgetq_lane_s32): Delete.
> (vgetq_lane_s8): Delete.
> (vgetq_lane_s64): Delete.
> (vgetq_lane_u8): Delete.
> (vgetq_lane_u16): Delete.
> (vgetq_lane_u32): Delete.
> (vgetq_lane_u64): Delete.
> (__ARM_NUM_LANES): Delete.
> (__ARM_LANEQ): Delete.
> (__ARM_CHECK_LANEQ): Delete.
> (__arm_vsetq_lane_s16): Delete.
> (__arm_vsetq_lane_s32): Delete.
> (__arm_vsetq_lane_s8): Delete.
> (__arm_vsetq_lane_s64): Delete.
> (__arm_vsetq_lane_u8): Delete.
> (__arm_vsetq_lane_u16): Delete.
> (__arm_vsetq_lane_u32): Delete.
> (__arm_vsetq_lane_u64): Delete.
> (__arm_vgetq_lane_s16): Delete.
> (__arm_vgetq_lane_s32): Delete.
> (__arm_vgetq_lane_s8): Delete.
> (__arm_vgetq_lane_s64): Delete.
> (__arm_vgetq_lane_u8): Delete.
> (__arm_vgetq_lane_u16): Delete.
> (__arm_vgetq_lane_u32): Delete.
> (__arm_vgetq_lane_u64): Delete.
> (__arm_vsetq_lane_f16): Delete.
> (__arm_vsetq_lane_f32): Delete.
> (__arm_vgetq_lane_f16): Delete.
> (__arm_vgetq_lane_f32): Delete.
> (__arm_vsetq_lane): Delete.
> (__arm_vgetq_lane): Delete.
> * config/arm/mve.md (mve_vec_extract<mode><V_elem_l>): Add '@'
> prefix.
> (mve_vec_set<mode>_internal): Likewise.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/arm/mve/intrinsics/vsetq_lane_f16_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_f32_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_s16_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_s32_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_s64_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_s8_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_u16_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_u32_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_u64_bounds.c: New test.
> * gcc.target/arm/mve/intrinsics/vsetq_lane_u8_bounds.c: New test.
> ---
> gcc/config/arm/arm-mve-builtins-base.cc | 59 +++
> gcc/config/arm/arm-mve-builtins-base.def | 4 +
> gcc/config/arm/arm-mve-builtins-base.h | 2 +
> gcc/config/arm/arm-mve-builtins-shapes.cc | 82 ++++
> gcc/config/arm/arm-mve-builtins-shapes.h | 2 +
> gcc/config/arm/arm_mve.h | 401 ------------------
> gcc/config/arm/mve.md | 5 +-
> .../mve/intrinsics/vsetq_lane_f16_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_f32_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_s16_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_s32_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_s64_bounds.c | 19 +
> .../arm/mve/intrinsics/vsetq_lane_s8_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_u16_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_u32_bounds.c | 19 +
> .../mve/intrinsics/vsetq_lane_u64_bounds.c | 19 +
> .../arm/mve/intrinsics/vsetq_lane_u8_bounds.c | 19 +
> 17 files changed, 342 insertions(+), 403 deletions(-)
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f16_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f32_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s16_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s32_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s64_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s8_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u16_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u32_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u64_bounds.c
> create mode 100644
> gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u8_bounds.c
>
> diff --git a/gcc/config/arm/arm-mve-builtins-base.cc
> b/gcc/config/arm/arm-mve-builtins-base.cc
> index fd0be2cbcd8..f894f163d2e 100644
> --- a/gcc/config/arm/arm-mve-builtins-base.cc
> +++ b/gcc/config/arm/arm-mve-builtins-base.cc
> @@ -1167,6 +1167,63 @@ public:
> }
> };
>
> + /* Map the function directly to mve_vec_set_internal (M) or mve_vec_extract
> + (M, M) where M is the vector mode associated with type suffix 0, except
> when
> + mode is V2DI where the builtin name is hardcoded. */
> +class mve_function_vsetq_vgetq_lane : public function_base
> +{
> +public:
> + CONSTEXPR mve_function_vsetq_vgetq_lane (bool is_get)
> + : m_is_get (is_get)
> + {}
> +
> + /* True for vgetq_lane, false for vsetq_lane. */
> + bool m_is_get;
> +
> + rtx
> + expand (function_expander &e) const override
> + {
> + machine_mode mode = e.vector_mode (0);
> + insn_code code;
> + HOST_WIDE_INT elem;
> +
> + code = (mode == V2DImode)
> + ? (m_is_get
> + ? CODE_FOR_mve_vec_extractv2didi
> + : CODE_FOR_mve_vec_setv2di_internal)
> + : (m_is_get
> + ? code_for_mve_vec_extract (mode, mode)
> + : code_for_mve_vec_set_internal (mode));
> +
> + if (!m_is_get)
> + {
> + /* mve_vec_set has vector and lane number arguments in opposite order
> + compared to the intrinsic: swap them now... */
> + std::swap (e.args[1], e.args[2]);
> + }
> +
> + elem = INTVAL (e.args[1]);
> +
> + /* For big-endian, GCC's vector indices are reversed within each 64 bits
> + compared to the architectural lane indices used by MVE intrinsics. */
> + if (BYTES_BIG_ENDIAN)
> + {
> + unsigned int num_lanes = 128 / e.type_suffix (0).element_bits;
> + elem ^= (num_lanes / 2) - 1;
> + }
> +
> + if (!m_is_get)
> + {
> + /* ... and convert the lane number into a mask as expected by the
> + builtin. */
> + elem = HOST_WIDE_INT_1 << elem;
> + }
> + e.args[1] = GEN_INT (elem);
> +
> + return e.use_unpred_insn (code);
> + }
> +};
> +
> } /* end anonymous namespace */
>
> namespace arm_mve {
> @@ -1390,6 +1447,7 @@ FUNCTION_WITH_RTX_M (veorq, XOR, VEORQ)
> FUNCTION (vfmaq, unspec_mve_function_exact_insn, (-1, -1, VFMAQ_F, -1, -1,
> VFMAQ_N_F, -1, -1, VFMAQ_M_F, -1, -1, VFMAQ_M_N_F))
> FUNCTION (vfmasq, unspec_mve_function_exact_insn, (-1, -1, -1, -1, -1,
> VFMASQ_N_F, -1, -1, -1, -1, -1, VFMASQ_M_N_F))
> FUNCTION (vfmsq, unspec_mve_function_exact_insn, (-1, -1, VFMSQ_F, -1, -1,
> -1, -1, -1, VFMSQ_M_F, -1, -1, -1))
> +FUNCTION (vgetq_lane, mve_function_vsetq_vgetq_lane, (true))
> FUNCTION_WITH_M_N_NO_F (vhaddq, VHADDQ)
> FUNCTION_WITH_M_N_NO_F (vhsubq, VHSUBQ)
> FUNCTION (vld1q, vld1_impl,)
> @@ -1517,6 +1575,7 @@ FUNCTION_ONLY_N_NO_F (vrshrntq, VRSHRNTQ)
> FUNCTION_ONLY_N_NO_F (vrshrq, VRSHRQ)
> FUNCTION (vsbciq, vadc_vsbc_impl, (true, false))
> FUNCTION (vsbcq, vadc_vsbc_impl, (false, false))
> +FUNCTION (vsetq_lane, mve_function_vsetq_vgetq_lane, (false))
> FUNCTION (vshlcq, vshlc_impl,)
> FUNCTION_ONLY_N_NO_F (vshllbq, VSHLLBQ)
> FUNCTION_ONLY_N_NO_F (vshlltq, VSHLLTQ)
> diff --git a/gcc/config/arm/arm-mve-builtins-base.def
> b/gcc/config/arm/arm-mve-builtins-base.def
> index 0400c3c540e..51a708957e5 100644
> --- a/gcc/config/arm/arm-mve-builtins-base.def
> +++ b/gcc/config/arm/arm-mve-builtins-base.def
> @@ -52,6 +52,7 @@ DEF_MVE_FUNCTION (vddupq, viddup, all_unsigned, mx_or_none)
> DEF_MVE_FUNCTION (vdupq, unary_n, all_integer, mx_or_none)
> DEF_MVE_FUNCTION (vdwdupq, vidwdup, all_unsigned, mx_or_none)
> DEF_MVE_FUNCTION (veorq, binary, all_integer, mx_or_none)
> +DEF_MVE_FUNCTION (vgetq_lane, getq_lane, all_integer_with_64, none)
> DEF_MVE_FUNCTION (vhaddq, binary_opt_n, all_integer, mx_or_none)
> DEF_MVE_FUNCTION (vhcaddq_rot270, binary, all_signed, mx_or_none)
> DEF_MVE_FUNCTION (vhcaddq_rot90, binary, all_signed, mx_or_none)
> @@ -171,6 +172,7 @@ DEF_MVE_FUNCTION (vrshrq, binary_rshift, all_integer,
> mx_or_none)
> DEF_MVE_FUNCTION (vsbciq, vadc_vsbc, integer_32, m_or_none)
> DEF_MVE_FUNCTION (vsbcq, vadc_vsbc, integer_32, m_or_none)
> DEF_MVE_FUNCTION (vshlcq, vshlc, all_integer, m_or_none)
> +DEF_MVE_FUNCTION (vsetq_lane, setq_lane, all_integer_with_64, none)
> DEF_MVE_FUNCTION (vshllbq, binary_widen_n, integer_8_16, mx_or_none)
> DEF_MVE_FUNCTION (vshlltq, binary_widen_n, integer_8_16, mx_or_none)
> DEF_MVE_FUNCTION (vshlq, binary_lshift, all_integer, mx_or_none)
> @@ -237,6 +239,7 @@ DEF_MVE_FUNCTION (veorq, binary, all_float, mx_or_none)
> DEF_MVE_FUNCTION (vfmaq, ternary_opt_n, all_float, m_or_none)
> DEF_MVE_FUNCTION (vfmasq, ternary_n, all_float, m_or_none)
> DEF_MVE_FUNCTION (vfmsq, ternary, all_float, m_or_none)
> +DEF_MVE_FUNCTION (vgetq_lane, getq_lane, all_float, none)
> DEF_MVE_FUNCTION (vld1q, load, all_float, z_or_none)
> DEF_MVE_FUNCTION (vld2q, load, all_float, none)
> DEF_MVE_FUNCTION (vld4q, load, all_float, none)
> @@ -269,6 +272,7 @@ DEF_MVE_FUNCTION (vrndnq, unary, all_float, mx_or_none)
> DEF_MVE_FUNCTION (vrndpq, unary, all_float, mx_or_none)
> DEF_MVE_FUNCTION (vrndq, unary, all_float, mx_or_none)
> DEF_MVE_FUNCTION (vrndxq, unary, all_float, mx_or_none)
> +DEF_MVE_FUNCTION (vsetq_lane, setq_lane, all_float, none)
> DEF_MVE_FUNCTION (vst1q, store, all_float, p_or_none)
> DEF_MVE_FUNCTION (vst2q, store, all_float, none)
> DEF_MVE_FUNCTION (vst4q, store, all_float, none)
> diff --git a/gcc/config/arm/arm-mve-builtins-base.h
> b/gcc/config/arm/arm-mve-builtins-base.h
> index 6ff3195e888..da5ad0c8aaa 100644
> --- a/gcc/config/arm/arm-mve-builtins-base.h
> +++ b/gcc/config/arm/arm-mve-builtins-base.h
> @@ -75,6 +75,7 @@ extern const function_base *const veorq;
> extern const function_base *const vfmaq;
> extern const function_base *const vfmasq;
> extern const function_base *const vfmsq;
> +extern const function_base *const vgetq_lane;
> extern const function_base *const vhaddq;
> extern const function_base *const vhcaddq_rot270;
> extern const function_base *const vhcaddq_rot90;
> @@ -206,6 +207,7 @@ extern const function_base *const vrshrntq;
> extern const function_base *const vrshrq;
> extern const function_base *const vsbciq;
> extern const function_base *const vsbcq;
> +extern const function_base *const vsetq_lane;
> extern const function_base *const vshlcq;
> extern const function_base *const vshllbq;
> extern const function_base *const vshlltq;
> diff --git a/gcc/config/arm/arm-mve-builtins-shapes.cc
> b/gcc/config/arm/arm-mve-builtins-shapes.cc
> index aeb14530a08..4b8cc244448 100644
> --- a/gcc/config/arm/arm-mve-builtins-shapes.cc
> +++ b/gcc/config/arm/arm-mve-builtins-shapes.cc
> @@ -1467,6 +1467,47 @@ struct create_def : public nonoverloaded_base
> };
> SHAPE (create)
>
> +/* <S0>_t vfoo[_t0](<T0>_t, const int)
> +
> + Check that 'idx' is in the [0..#num_lanes - 1] range.
> +
> + Example: vgetq_lane.
> + int8_t [__arm_]vgetq_lane[_s8](int8x16_t a, const int idx) */
> +
> +struct getq_lane_def : public overloaded_base<0>
> +{
> + void
> + build (function_builder &b, const function_group_info &group,
> + bool preserve_user_namespace) const override
> + {
> + b.add_overloaded_functions (group, MODE_none, preserve_user_namespace);
> + build_all (b, "s0,v0,su64", group, MODE_none, preserve_user_namespace);
> + }
> +
> + tree
> + resolve (function_resolver &r) const override
> + {
> + unsigned int i, nargs;
> + type_suffix_index type;
> + if (!r.check_gp_argument (2, i, nargs)
> + || (type = r.infer_vector_type (i-1)) == NUM_TYPE_SUFFIXES
> + || !r.require_integer_immediate (i))
> + return error_mark_node;
> +
> + return r.resolve_to (r.mode_suffix_id, type);
> + }
> +
> + bool
> + check (function_checker &c) const override
> + {
> + unsigned int num_lanes = 128 / c.type_suffix (0).element_bits;
> +
> + return c.require_immediate_range (1, 0, num_lanes - 1);
> + }
> +
> +};
> +SHAPE (getq_lane)
> +
> /* <T0>[xN]_t vfoo_t0().
>
> Example: vuninitializedq.
> @@ -1682,6 +1723,47 @@ struct mvn_def : public overloaded_base<0>
> };
> SHAPE (mvn)
>
> +/* <T0>_t vfoo[_t0](<S0>_t, <T0>_t, const_int)
> +
> + Check that 'idx' is in the [0..#num_lanes - 1] range.
> +
> + Example: vsetq_lane.
> + int8x16_t [__arm_]vsetq_lane[_s8](int8_t a, int8x16_t b, const int idx)
> */
> +struct setq_lane_def : public overloaded_base<0>
> +{
> + void
> + build (function_builder &b, const function_group_info &group,
> + bool preserve_user_namespace) const override
> + {
> + b.add_overloaded_functions (group, MODE_none, preserve_user_namespace);
> + build_all (b, "v0,s0,v0,su64", group, MODE_none,
> preserve_user_namespace);
> + }
> +
> + tree
> + resolve (function_resolver &r) const override
> + {
> + unsigned int i, nargs;
> + type_suffix_index type;
> + if (!r.check_gp_argument (3, i, nargs)
> + || (type = r.infer_vector_type (i-1)) == NUM_TYPE_SUFFIXES
> + || !r.require_derived_scalar_type (i - 2, r.SAME_TYPE_CLASS)
> + || !r.require_integer_immediate (i))
> + return error_mark_node;
> +
> + return r.resolve_to (r.mode_suffix_id, type);
> + }
> +
> + bool
> + check (function_checker &c) const override
> + {
> + unsigned int num_lanes = 128 / c.type_suffix (0).element_bits;
> +
> + return c.require_immediate_range (2, 0, num_lanes - 1);
> + }
> +
> +};
> +SHAPE (setq_lane)
> +
> /* void vfoo[_t0](<X>_t *, <T0>[xN]_t)
>
> where <X> might be tied to <t0> (for non-truncating stores) or might
> diff --git a/gcc/config/arm/arm-mve-builtins-shapes.h
> b/gcc/config/arm/arm-mve-builtins-shapes.h
> index 6b839c577c6..9b1d32a0056 100644
> --- a/gcc/config/arm/arm-mve-builtins-shapes.h
> +++ b/gcc/config/arm/arm-mve-builtins-shapes.h
> @@ -60,12 +60,14 @@ namespace arm_mve
> extern const function_shape *const binary_widen_poly;
> extern const function_shape *const cmp;
> extern const function_shape *const create;
> + extern const function_shape *const getq_lane;
> extern const function_shape *const inherent;
> extern const function_shape *const load;
> extern const function_shape *const load_ext;
> extern const function_shape *const load_ext_gather_offset;
> extern const function_shape *const load_gather_base;
> extern const function_shape *const mvn;
> + extern const function_shape *const setq_lane;
> extern const function_shape *const store;
> extern const function_shape *const store_scatter_base;
> extern const function_shape *const store_scatter_offset;
> diff --git a/gcc/config/arm/arm_mve.h b/gcc/config/arm/arm_mve.h
> index ee18a4714fb..7fc1cc3173f 100644
> --- a/gcc/config/arm/arm_mve.h
> +++ b/gcc/config/arm/arm_mve.h
> @@ -46,8 +46,6 @@
>
> #ifndef __ARM_MVE_PRESERVE_USER_NAMESPACE
> #define vuninitializedq(__v) __arm_vuninitializedq(__v)
> -#define vsetq_lane(__a, __b, __idx) __arm_vsetq_lane(__a, __b, __idx)
> -#define vgetq_lane(__a, __idx) __arm_vgetq_lane(__a, __idx)
>
>
> #define vpnot(__a) __arm_vpnot(__a)
> @@ -61,26 +59,6 @@
> #define vuninitializedq_s64(void) __arm_vuninitializedq_s64(void)
> #define vuninitializedq_f16(void) __arm_vuninitializedq_f16(void)
> #define vuninitializedq_f32(void) __arm_vuninitializedq_f32(void)
> -#define vsetq_lane_f16(__a, __b, __idx) __arm_vsetq_lane_f16(__a, __b,
> __idx)
> -#define vsetq_lane_f32(__a, __b, __idx) __arm_vsetq_lane_f32(__a, __b,
> __idx)
> -#define vsetq_lane_s16(__a, __b, __idx) __arm_vsetq_lane_s16(__a, __b,
> __idx)
> -#define vsetq_lane_s32(__a, __b, __idx) __arm_vsetq_lane_s32(__a, __b,
> __idx)
> -#define vsetq_lane_s8(__a, __b, __idx) __arm_vsetq_lane_s8(__a, __b, __idx)
> -#define vsetq_lane_s64(__a, __b, __idx) __arm_vsetq_lane_s64(__a, __b,
> __idx)
> -#define vsetq_lane_u8(__a, __b, __idx) __arm_vsetq_lane_u8(__a, __b, __idx)
> -#define vsetq_lane_u16(__a, __b, __idx) __arm_vsetq_lane_u16(__a, __b,
> __idx)
> -#define vsetq_lane_u32(__a, __b, __idx) __arm_vsetq_lane_u32(__a, __b,
> __idx)
> -#define vsetq_lane_u64(__a, __b, __idx) __arm_vsetq_lane_u64(__a, __b,
> __idx)
> -#define vgetq_lane_f16(__a, __idx) __arm_vgetq_lane_f16(__a, __idx)
> -#define vgetq_lane_f32(__a, __idx) __arm_vgetq_lane_f32(__a, __idx)
> -#define vgetq_lane_s16(__a, __idx) __arm_vgetq_lane_s16(__a, __idx)
> -#define vgetq_lane_s32(__a, __idx) __arm_vgetq_lane_s32(__a, __idx)
> -#define vgetq_lane_s8(__a, __idx) __arm_vgetq_lane_s8(__a, __idx)
> -#define vgetq_lane_s64(__a, __idx) __arm_vgetq_lane_s64(__a, __idx)
> -#define vgetq_lane_u8(__a, __idx) __arm_vgetq_lane_u8(__a, __idx)
> -#define vgetq_lane_u16(__a, __idx) __arm_vgetq_lane_u16(__a, __idx)
> -#define vgetq_lane_u32(__a, __idx) __arm_vgetq_lane_u32(__a, __idx)
> -#define vgetq_lane_u64(__a, __idx) __arm_vgetq_lane_u64(__a, __idx)
> #define sqrshr(__p0, __p1) __arm_sqrshr(__p0, __p1)
> #define sqrshrl(__p0, __p1) __arm_sqrshrl(__p0, __p1)
> #define sqrshrl_sat48(__p0, __p1) __arm_sqrshrl_sat48(__p0, __p1)
> @@ -99,18 +77,6 @@
> #define asrl(__p0, __p1) __arm_asrl(__p0, __p1)
> #endif
>
> -/* For big-endian, GCC's vector indices are reversed within each 64 bits
> - compared to the architectural lane indices used by MVE intrinsics. */
> -#define __ARM_NUM_LANES(__v) (sizeof (__v) / sizeof (__v[0]))
> -#ifdef __ARM_BIG_ENDIAN
> -#define __ARM_LANEQ(__vec, __idx) (__idx ^ (__ARM_NUM_LANES(__vec)/2 - 1))
> -#else
> -#define __ARM_LANEQ(__vec, __idx) __idx
> -#endif
> -#define __ARM_CHECK_LANEQ(__vec, __idx) \
> - __builtin_arm_lane_check (__ARM_NUM_LANES(__vec), \
> - __ARM_LANEQ(__vec, __idx))
> -
> __extension__ extern __inline mve_pred16_t
> __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> __arm_vpnot (mve_pred16_t __a)
> @@ -118,142 +84,6 @@ __arm_vpnot (mve_pred16_t __a)
> return __builtin_mve_vpnotv16bi (__a);
> }
>
> -__extension__ extern __inline int16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_s16 (int16_t __a, int16x8_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline int32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_s32 (int32_t __a, int32x4_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline int8x16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_s8 (int8_t __a, int8x16_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline int64x2_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_s64 (int64_t __a, int64x2_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline uint8x16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_u8 (uint8_t __a, uint8x16_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline uint16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_u16 (uint16_t __a, uint16x8_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline uint32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_u32 (uint32_t __a, uint32x4_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline uint64x2_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_u64 (uint64_t __a, uint64x2_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline int16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_s16 (int16x8_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline int32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_s32 (int32x4_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline int8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_s8 (int8x16_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline int64_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_s64 (int64x2_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline uint8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_u8 (uint8x16_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline uint16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_u16 (uint16x8_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline uint32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_u32 (uint32x4_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline uint64_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_u64 (uint64x2_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> __extension__ extern __inline uint64_t
> __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> __arm_lsll (uint64_t value, int32_t shift)
> @@ -366,189 +196,8 @@ __arm_srshr (int32_t value, const int shift)
> return __builtin_mve_srshr_si (value, shift);
> }
>
> -#if (__ARM_FEATURE_MVE & 2) /* MVE Floating point. */
> -
> -__extension__ extern __inline float16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_f16 (float16_t __a, float16x8_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline float32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane_f32 (float32_t __a, float32x4_t __b, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__b, __idx);
> - __b[__ARM_LANEQ(__b,__idx)] = __a;
> - return __b;
> -}
> -
> -__extension__ extern __inline float16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_f16 (float16x8_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -
> -__extension__ extern __inline float32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane_f32 (float32x4_t __a, const int __idx)
> -{
> - __ARM_CHECK_LANEQ (__a, __idx);
> - return __a[__ARM_LANEQ(__a,__idx)];
> -}
> -#endif
> -
> #ifdef __cplusplus
>
> -__extension__ extern __inline int16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (int16_t __a, int16x8_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_s16 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline int32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (int32_t __a, int32x4_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_s32 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline int8x16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (int8_t __a, int8x16_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_s8 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline int64x2_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (int64_t __a, int64x2_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_s64 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline uint8x16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (uint8_t __a, uint8x16_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_u8 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline uint16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (uint16_t __a, uint16x8_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_u16 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline uint32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (uint32_t __a, uint32x4_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_u32 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline uint64x2_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (uint64_t __a, uint64x2_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_u64 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline int16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (int16x8_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_s16 (__a, __idx);
> -}
> -
> -__extension__ extern __inline int32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (int32x4_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_s32 (__a, __idx);
> -}
> -
> -__extension__ extern __inline int8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (int8x16_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_s8 (__a, __idx);
> -}
> -
> -__extension__ extern __inline int64_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (int64x2_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_s64 (__a, __idx);
> -}
> -
> -__extension__ extern __inline uint8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (uint8x16_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_u8 (__a, __idx);
> -}
> -
> -__extension__ extern __inline uint16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (uint16x8_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_u16 (__a, __idx);
> -}
> -
> -__extension__ extern __inline uint32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (uint32x4_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_u32 (__a, __idx);
> -}
> -
> -__extension__ extern __inline uint64_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (uint64x2_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_u64 (__a, __idx);
> -}
> -
> -#if (__ARM_FEATURE_MVE & 2) /* MVE Floating point. */
> -
> -__extension__ extern __inline float16x8_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (float16_t __a, float16x8_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_f16 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline float32x4_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vsetq_lane (float32_t __a, float32x4_t __b, const int __idx)
> -{
> - return __arm_vsetq_lane_f32 (__a, __b, __idx);
> -}
> -
> -__extension__ extern __inline float16_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (float16x8_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_f16 (__a, __idx);
> -}
> -
> -__extension__ extern __inline float32_t
> -__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> -__arm_vgetq_lane (float32x4_t __a, const int __idx)
> -{
> - return __arm_vgetq_lane_f32 (__a, __idx);
> -}
> -#endif /* MVE Floating point. */
> -
> -
> __extension__ extern __inline uint8x16_t
> __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> __arm_vuninitializedq (uint8x16_t /* __v ATTRIBUTE UNUSED */)
> @@ -871,33 +520,6 @@ extern void *__ARM_undef;
> int (*)[__ARM_mve_type_float16x8_t]: __arm_vuninitializedq_f16 (), \
> int (*)[__ARM_mve_type_float32x4_t]: __arm_vuninitializedq_f32 ());})
>
> -#define __arm_vgetq_lane(p0,p1) ({ __typeof(p0) __p0 = (p0); \
> - _Generic( (int (*)[__ARM_mve_typeid(__p0)])0, \
> - int (*)[__ARM_mve_type_int8x16_t]: __arm_vgetq_lane_s8
> (__ARM_mve_coerce(__p0, int8x16_t), p1), \
> - int (*)[__ARM_mve_type_int16x8_t]: __arm_vgetq_lane_s16
> (__ARM_mve_coerce(__p0, int16x8_t), p1), \
> - int (*)[__ARM_mve_type_int32x4_t]: __arm_vgetq_lane_s32
> (__ARM_mve_coerce(__p0, int32x4_t), p1), \
> - int (*)[__ARM_mve_type_int64x2_t]: __arm_vgetq_lane_s64
> (__ARM_mve_coerce(__p0, int64x2_t), p1), \
> - int (*)[__ARM_mve_type_uint8x16_t]: __arm_vgetq_lane_u8
> (__ARM_mve_coerce(__p0, uint8x16_t), p1), \
> - int (*)[__ARM_mve_type_uint16x8_t]: __arm_vgetq_lane_u16
> (__ARM_mve_coerce(__p0, uint16x8_t), p1), \
> - int (*)[__ARM_mve_type_uint32x4_t]: __arm_vgetq_lane_u32
> (__ARM_mve_coerce(__p0, uint32x4_t), p1), \
> - int (*)[__ARM_mve_type_uint64x2_t]: __arm_vgetq_lane_u64
> (__ARM_mve_coerce(__p0, uint64x2_t), p1), \
> - int (*)[__ARM_mve_type_float16x8_t]: __arm_vgetq_lane_f16
> (__ARM_mve_coerce(__p0, float16x8_t), p1), \
> - int (*)[__ARM_mve_type_float32x4_t]: __arm_vgetq_lane_f32
> (__ARM_mve_coerce(__p0, float32x4_t), p1));})
> -
> -#define __arm_vsetq_lane(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
> - __typeof(p1) __p1 = (p1); \
> - _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)])0, \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int8x16_t]:
> __arm_vsetq_lane_s8 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int8x16_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int16x8_t]:
> __arm_vsetq_lane_s16 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int16x8_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int32x4_t]:
> __arm_vsetq_lane_s32 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int32x4_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int64x2_t]:
> __arm_vsetq_lane_s64 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int64x2_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint8x16_t]:
> __arm_vsetq_lane_u8 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint8x16_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint16x8_t]:
> __arm_vsetq_lane_u16 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint16x8_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint32x4_t]:
> __arm_vsetq_lane_u32 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint32x4_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint64x2_t]:
> __arm_vsetq_lane_u64 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint64x2_t), p2), \
> - int (*)[__ARM_mve_type_fp_n][__ARM_mve_type_float16x8_t]:
> __arm_vsetq_lane_f16 (__ARM_mve_coerce_f_scalar(__p0, double),
> __ARM_mve_coerce(__p1, float16x8_t), p2), \
> - int (*)[__ARM_mve_type_fp_n][__ARM_mve_type_float32x4_t]:
> __arm_vsetq_lane_f32 (__ARM_mve_coerce_f_scalar(__p0, double),
> __ARM_mve_coerce(__p1, float32x4_t), p2));})
> -
> #else /* MVE Integer. */
>
> #define __arm_vuninitializedq(p0) ({ __typeof(p0) __p0 = (p0); \
> @@ -911,29 +533,6 @@ extern void *__ARM_undef;
> int (*)[__ARM_mve_type_uint32x4_t]: __arm_vuninitializedq_u32 (), \
> int (*)[__ARM_mve_type_uint64x2_t]: __arm_vuninitializedq_u64 ());})
>
> -#define __arm_vgetq_lane(p0,p1) ({ __typeof(p0) __p0 = (p0); \
> - _Generic( (int (*)[__ARM_mve_typeid(__p0)])0, \
> - int (*)[__ARM_mve_type_int8x16_t]: __arm_vgetq_lane_s8
> (__ARM_mve_coerce(__p0, int8x16_t), p1), \
> - int (*)[__ARM_mve_type_int16x8_t]: __arm_vgetq_lane_s16
> (__ARM_mve_coerce(__p0, int16x8_t), p1), \
> - int (*)[__ARM_mve_type_int32x4_t]: __arm_vgetq_lane_s32
> (__ARM_mve_coerce(__p0, int32x4_t), p1), \
> - int (*)[__ARM_mve_type_int64x2_t]: __arm_vgetq_lane_s64
> (__ARM_mve_coerce(__p0, int64x2_t), p1), \
> - int (*)[__ARM_mve_type_uint8x16_t]: __arm_vgetq_lane_u8
> (__ARM_mve_coerce(__p0, uint8x16_t), p1), \
> - int (*)[__ARM_mve_type_uint16x8_t]: __arm_vgetq_lane_u16
> (__ARM_mve_coerce(__p0, uint16x8_t), p1), \
> - int (*)[__ARM_mve_type_uint32x4_t]: __arm_vgetq_lane_u32
> (__ARM_mve_coerce(__p0, uint32x4_t), p1), \
> - int (*)[__ARM_mve_type_uint64x2_t]: __arm_vgetq_lane_u64
> (__ARM_mve_coerce(__p0, uint64x2_t), p1));})
> -
> -#define __arm_vsetq_lane(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
> - __typeof(p1) __p1 = (p1); \
> - _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)])0, \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int8x16_t]:
> __arm_vsetq_lane_s8 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int8x16_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int16x8_t]:
> __arm_vsetq_lane_s16 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int16x8_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int32x4_t]:
> __arm_vsetq_lane_s32 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int32x4_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_int64x2_t]:
> __arm_vsetq_lane_s64 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, int64x2_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint8x16_t]:
> __arm_vsetq_lane_u8 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint8x16_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint16x8_t]:
> __arm_vsetq_lane_u16 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint16x8_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint32x4_t]:
> __arm_vsetq_lane_u32 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint32x4_t), p2), \
> - int (*)[__ARM_mve_type_int_n][__ARM_mve_type_uint64x2_t]:
> __arm_vsetq_lane_u64 (__ARM_mve_coerce_i_scalar(__p0, int),
> __ARM_mve_coerce(__p1, uint64x2_t), p2));})
> -
> #endif /* MVE Integer. */
>
> #endif /* __cplusplus */
> diff --git a/gcc/config/arm/mve.md b/gcc/config/arm/mve.md
> index 8527bd753e3..70b1c77fdce 100644
> --- a/gcc/config/arm/mve.md
> +++ b/gcc/config/arm/mve.md
> @@ -4146,10 +4146,11 @@ (define_insn "@mve_vld4q<mode>"
> return "";
> }
> [(set_attr "length" "16")])
> +
> ;;
> ;; [vgetq_lane_u, vgetq_lane_s, vgetq_lane_f])
> ;;
> -(define_insn "mve_vec_extract<mode><V_elem_l>"
> +(define_insn "@mve_vec_extract<mode><V_elem_l>"
> [(set (match_operand:<V_elem> 0 "nonimmediate_operand" "=r")
> (vec_select:<V_elem>
> (match_operand:MVE_VLD_ST 1 "s_register_operand" "w")
> @@ -4224,7 +4225,7 @@ (define_insn "*mve_vec_extract_zext_internal<mode>"
> ;;
> ;; [vsetq_lane_u, vsetq_lane_s, vsetq_lane_f])
> ;;
> -(define_insn "mve_vec_set<mode>_internal"
> +(define_insn "@mve_vec_set<mode>_internal"
> [(set (match_operand:VQ2 0 "s_register_operand" "=w")
> (vec_merge:VQ2
> (vec_duplicate:VQ2
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f16_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f16_bounds.c
> new file mode 100644
> index 00000000000..0aec13fdf4e
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f16_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_fp_ok } */
> +/* { dg-add-options arm_v8_1m_mve_fp } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +float16x8_t
> +foo (float16_t a, float16x8_t b)
> +{
> + return vsetq_lane_f16 (a, b, 9); /* { dg-error {passing 9 to argument 3 of
> 'vsetq_lane_f16', which expects a value in the range \[0, 7\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f32_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f32_bounds.c
> new file mode 100644
> index 00000000000..5627e0a17c1
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_f32_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_fp_ok } */
> +/* { dg-add-options arm_v8_1m_mve_fp } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +float32x4_t
> +foo (float32_t a, float32x4_t b)
> +{
> + return vsetq_lane_f32 (a, b, 4); /* { dg-error {passing 4 to argument 3 of
> 'vsetq_lane_f32', which expects a value in the range \[0, 3\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s16_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s16_bounds.c
> new file mode 100644
> index 00000000000..af255f8dba9
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s16_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +int16x8_t
> +foo (int16_t a, int16x8_t b)
> +{
> + return vsetq_lane_s16 (a, b, 9); /* { dg-error {passing 9 to argument 3 of
> 'vsetq_lane_s16', which expects a value in the range \[0, 7\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s32_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s32_bounds.c
> new file mode 100644
> index 00000000000..968910cf0a3
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s32_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +int32x4_t
> +foo (int32_t a, int32x4_t b)
> +{
> + return vsetq_lane_s32 (a, b, 4); /* { dg-error {passing 4 to argument 3 of
> 'vsetq_lane_s32', which expects a value in the range \[0, 3\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s64_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s64_bounds.c
> new file mode 100644
> index 00000000000..f3bf3ac3886
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s64_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +int64x2_t
> +foo (int64_t a, int64x2_t b)
> +{
> + return vsetq_lane_s64 (a, b, 2); /* { dg-error {passing 2 to argument 3 of
> 'vsetq_lane_s64', which expects a value in the range \[0, 1\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s8_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s8_bounds.c
> new file mode 100644
> index 00000000000..0f4f2380555
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_s8_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +int8x16_t
> +foo (int8_t a, int8x16_t b)
> +{
> + return vsetq_lane_s8 (a, b, 17); /* { dg-error {passing 17 to argument 3
> of 'vsetq_lane_s8', which expects a value in the range \[0, 15\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u16_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u16_bounds.c
> new file mode 100644
> index 00000000000..62a03c0dc50
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u16_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +uint16x8_t
> +foo (uint16_t a, uint16x8_t b)
> +{
> + return vsetq_lane_u16 (a, b, 8); /* { dg-error {passing 8 to argument 3 of
> 'vsetq_lane_u16', which expects a value in the range \[0, 7\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u32_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u32_bounds.c
> new file mode 100644
> index 00000000000..5836eeb994f
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u32_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +uint32x4_t
> +foo (uint32_t a, uint32x4_t b)
> +{
> + return vsetq_lane_u32 (a, b, 4); /* { dg-error {passing 4 to argument 3 of
> 'vsetq_lane_u32', which expects a value in the range \[0, 3\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u64_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u64_bounds.c
> new file mode 100644
> index 00000000000..8be2b80e537
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u64_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +uint64x2_t
> +foo (uint64_t a, uint64x2_t b)
> +{
> + return vsetq_lane_u64 (a, b, 2); /* { dg-error {passing 2 to argument 3 of
> 'vsetq_lane_u64', which expects a value in the range \[0, 1\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> diff --git
> a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u8_bounds.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u8_bounds.c
> new file mode 100644
> index 00000000000..987dc98b5e8
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vsetq_lane_u8_bounds.c
> @@ -0,0 +1,19 @@
> +/* { dg-require-effective-target arm_v8_1m_mve_ok } */
> +/* { dg-add-options arm_v8_1m_mve } */
> +/* { dg-additional-options "-O2" } */
> +
> +#include "arm_mve.h"
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +uint8x16_t
> +foo (uint8_t a, uint8x16_t b)
> +{
> + return vsetq_lane_u8 (a, b, 17); /* { dg-error {passing 17 to argument 3
> of 'vsetq_lane_u8', which expects a value in the range \[0, 15\]} } */
> +}
> +
> +#ifdef __cplusplus
> +}
> +#endif
> --
> 2.34.1
>
