maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/SemaCXX/aarch64-sve-explicit-casts-...

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

53 lines
2.3 KiB
C++
Raw Normal View History

[AArch64][Driver][SVE] Allow -msve-vector-bits=<n>+ syntax to mean no maximum vscale This patch splits the existing SveVectorBits LangOpt into VScaleMin and VScaleMax LangOpts such that we can represent such an option. The cc1 option has also been split into -mvscale-{min,max}=<n> options so that the cc1 arguments better reflect the vscale_range IR attribute. Differential Revision: https://reviews.llvm.org/D111790
2021-10-14 19:04:33 +08:00
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -mvscale-min=1 -mvscale-max=1 -flax-vector-conversions=none -fallow-half-arguments-and-returns -ffreestanding -fsyntax-only -verify %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -mvscale-min=2 -mvscale-max=2 -flax-vector-conversions=none -fallow-half-arguments-and-returns -ffreestanding -fsyntax-only -verify %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -mvscale-min=4 -mvscale-max=4 -flax-vector-conversions=none -fallow-half-arguments-and-returns -ffreestanding -fsyntax-only -verify %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -mvscale-min=8 -mvscale-max=8 -flax-vector-conversions=none -fallow-half-arguments-and-returns -ffreestanding -fsyntax-only -verify %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -mvscale-min=16 -mvscale-max=16 -flax-vector-conversions=none -fallow-half-arguments-and-returns -ffreestanding -fsyntax-only -verify %s
[AArch64][SVE] Allow C-style casts between fixed-size and scalable vectors This patch allows C-style casting between fixed-size and scalable vectors. This kind of cast was previously blocked by the compiler, but it should be allowed. Differential Revision: https://reviews.llvm.org/D91262
2020-11-17 20:33:56 +08:00
// expected-no-diagnostics
#include <arm_sve.h>
#define N __ARM_FEATURE_SVE_BITS
#define FIXED_ATTR __attribute__((arm_sve_vector_bits(N)))
typedef svfloat32_t fixed_float32_t FIXED_ATTR;
typedef svfloat64_t fixed_float64_t FIXED_ATTR;
typedef svint32_t fixed_int32_t FIXED_ATTR;
typedef svint64_t fixed_int64_t FIXED_ATTR;
[AArch64][SVE] Handle svbool_t VLST <-> VLAT/GNUT conversion According to https://godbolt.org/z/q5rME1naY and acle, we found that there are different SVE conversion behaviours between clang and gcc. It turns out that llvm does not handle SVE predicates width properly. This patch 1) checks SVE predicates width rightly with svbool_t type. 2) removes warning on svbool_t VLST <-> VLAT/GNUT conversion. 3) disables VLST <-> VLAT/GNUT conversion between SVE vectors and predicates due to different width. Differential Revision: https://reviews.llvm.org/D106333
2021-07-19 21:10:17 +08:00
typedef svbool_t fixed_bool_t FIXED_ATTR;
[AArch64][SVE] Allow C-style casts between fixed-size and scalable vectors This patch allows C-style casting between fixed-size and scalable vectors. This kind of cast was previously blocked by the compiler, but it should be allowed. Differential Revision: https://reviews.llvm.org/D91262
2020-11-17 20:33:56 +08:00
// SVE VLSTs can be cast to SVE VLATs, regardless of lane size.
// NOTE: the list below is NOT exhaustive for all SVE types.
#define CAST(from, to) \
void from##_to_##to(from a, to b) { \
b = (to) a; \
}
#define TESTCASE(ty1, ty2) \
CAST(ty1, ty2) \
CAST(ty2, ty1)
TESTCASE(fixed_float32_t, svfloat32_t)
TESTCASE(fixed_float32_t, svfloat64_t)
TESTCASE(fixed_float32_t, svint32_t)
TESTCASE(fixed_float32_t, svint64_t)
TESTCASE(fixed_float64_t, svfloat32_t)
TESTCASE(fixed_float64_t, svfloat64_t)
TESTCASE(fixed_float64_t, svint32_t)
TESTCASE(fixed_float64_t, svint64_t)
TESTCASE(fixed_int32_t, svfloat32_t)
TESTCASE(fixed_int32_t, svfloat64_t)
TESTCASE(fixed_int32_t, svint32_t)
TESTCASE(fixed_int32_t, svint64_t)
TESTCASE(fixed_int64_t, svfloat32_t)
TESTCASE(fixed_int64_t, svfloat64_t)
TESTCASE(fixed_int64_t, svint32_t)
TESTCASE(fixed_int64_t, svint64_t)
[AArch64][SVE] Handle svbool_t VLST <-> VLAT/GNUT conversion According to https://godbolt.org/z/q5rME1naY and acle, we found that there are different SVE conversion behaviours between clang and gcc. It turns out that llvm does not handle SVE predicates width properly. This patch 1) checks SVE predicates width rightly with svbool_t type. 2) removes warning on svbool_t VLST <-> VLAT/GNUT conversion. 3) disables VLST <-> VLAT/GNUT conversion between SVE vectors and predicates due to different width. Differential Revision: https://reviews.llvm.org/D106333
2021-07-19 21:10:17 +08:00
TESTCASE(fixed_bool_t, svbool_t)