diff options
Diffstat (limited to 'clang')
132 files changed, 4759 insertions, 346 deletions
diff --git a/clang/docs/UsersManual.rst b/clang/docs/UsersManual.rst index e82b16f..fb22ad3 100644 --- a/clang/docs/UsersManual.rst +++ b/clang/docs/UsersManual.rst @@ -11,7 +11,7 @@ Introduction ============ The Clang Compiler is an open-source compiler for the C family of -programming languages, aiming to be the best in class implementation of +programming languages, aiming to be the best-in-class implementation of these languages. Clang builds on the LLVM optimizer and code generator, allowing it to provide high-quality optimization and code generation support for many targets. For more general information, please see the @@ -56,7 +56,7 @@ migration from GCC to Clang. In most cases, code "just works". Clang also provides an alternative driver, :ref:`clang-cl`, that is designed to be compatible with the Visual C++ compiler, cl.exe. -In addition to language specific features, Clang has a variety of +In addition to language-specific features, Clang has a variety of features that depend on what CPU architecture or operating system is being compiled for. Please see the :ref:`Target-Specific Features and Limitations <target_features>` section for more details. @@ -299,7 +299,7 @@ output format of the diagnostics that it generates. This option controls the output format of the filename, line number, and column printed in diagnostic messages. The options, and their - affect on formatting a simple conversion diagnostic, follow: + effect on formatting a simple conversion diagnostic, follow: **clang** (default) :: @@ -360,7 +360,7 @@ output format of the diagnostics that it generates. t.c:3:11: warning: conversion specifies type 'char *' but the argument has type 'int' [-Wformat,Format String] This category can be used by clients that want to group diagnostics - by category, so it should be a high level category. We want dozens + by category, so it should be a high-level category. We want dozens of these, not hundreds or thousands of them. .. _opt_fsave-optimization-record: @@ -750,7 +750,7 @@ control the crash diagnostics. Disable auto-generation of preprocessed source files during a clang crash. - The -fno-crash-diagnostics flag can be helpful for speeding the process + The ``-fno-crash-diagnostics`` flag can be helpful for speeding the process of generating a delta reduced test case. .. option:: -fcrash-diagnostics-dir=<dir> @@ -779,7 +779,7 @@ Options to Emit Optimization Reports ------------------------------------ Optimization reports trace, at a high-level, all the major decisions -done by compiler transformations. For instance, when the inliner +made by compiler transformations. For instance, when the inliner decides to inline function ``foo()`` into ``bar()``, or the loop unroller decides to unroll a loop N times, or the vectorizer decides to vectorize a loop body. @@ -845,11 +845,11 @@ compilations steps. .. option:: -fproc-stat-report= - This option requests driver to print used memory and execution time of each + This option requests the driver to print used memory and execution time of each compilation step. The ``clang`` driver during execution calls different tools, like compiler, assembler, linker etc. With this option the driver reports total execution time, the execution time spent in user mode and peak memory - usage of each the called tool. Value of the option specifies where the report + usage of each called tool. Value of the option specifies where the report is sent to. If it specifies a regular file, the data are saved to this file in CSV format: @@ -869,7 +869,7 @@ compilations steps. * peak memory usage in Kb. It is possible to specify this option without any value. In this case statistics - are printed on standard output in human readable format: + are printed on standard output in human-readable format: .. code-block:: console @@ -884,7 +884,7 @@ compilations steps. You can also use environment variables to control the process statistics reporting. Setting ``CC_PRINT_PROC_STAT`` to ``1`` enables the feature, the report goes to - stdout in human readable format. + stdout in human-readable format. Setting ``CC_PRINT_PROC_STAT_FILE`` to a fully qualified file path makes it report process statistics to the given file in the CSV format. Specifying a relative path will likely lead to multiple files with the same name created in different @@ -922,7 +922,7 @@ Clang options that don't fit neatly into other categories. most filenames can be written to the file without any special formatting. Different Make tools will treat different sets of characters as "special" and use different conventions for telling the Make tool that the character - is actually part of the filename. Normally Clang uses backslash to "escape" + is actually part of the filename. Normally, Clang uses backslash to "escape" a special character, which is the convention used by GNU Make. The -MV option tells Clang to put double-quotes around the entire filename, which is the convention used by NMake and Jom. @@ -957,7 +957,7 @@ Configuration files Configuration files group command-line options and allow all of them to be specified just by referencing the configuration file. They may be used, for -example, to collect options required to tune compilation for particular +example, to collect options required to tune compilation for a particular target, such as ``-L``, ``-I``, ``-l``, ``--sysroot``, codegen options, etc. Configuration files can be either specified on the command line or loaded @@ -986,7 +986,7 @@ either during build or during runtime. At build time, use ``CLANG_CONFIG_FILE_USER_DIR`` and ``CLANG_CONFIG_FILE_SYSTEM_DIR``. At run time use the ``--config-user-dir=`` and ``--config-system-dir=`` command line options. Specifying config directories at runtime overrides the config -directories set at build time The first file found is used. It is an error if +directories set at build time. The first file found is used. It is an error if the required file cannot be found. The default configuration files are searched for in the same directories @@ -996,7 +996,7 @@ the ``--no-default-config`` flag. First, the algorithm searches for a configuration file named ``<triple>-<driver>.cfg`` where `triple` is the triple for the target being -built for, and `driver` is the name of the currently used driver. The algorithm +built, and `driver` is the name of the currently used driver. The algorithm first attempts to use the canonical name for the driver used, then falls back to the one found in the executable name. @@ -1047,7 +1047,7 @@ It is not an error if either of these files is not found. The configuration file consists of command-line options specified on one or more lines. Lines composed of whitespace characters only are ignored as well as lines in which the first non-blank character is ``#``. Long options may be split -between several lines by a trailing backslash. Here is example of a +between several lines by a trailing backslash. Here is an example of a configuration file: :: @@ -1229,7 +1229,7 @@ Clang also allows you to push and pop the current warning state. This is particularly useful when writing a header file that will be compiled by other people, because you don't know what warning flags they build with. -In the below example :option:`-Wextra-tokens` is ignored for only a single line +In the example below, :option:`-Wextra-tokens` is ignored for only a single line of code, after which the diagnostics return to whatever state had previously existed. @@ -1253,7 +1253,7 @@ of warnings, so even when using GCC-compatible #pragmas there is no guarantee that they will have identical behaviour on both compilers. Clang also doesn't yet support GCC behavior for ``#pragma diagnostic pop`` -that doesn't have a corresponding ``#pragma diagnostic push``. In this case +that doesn't have a corresponding ``#pragma diagnostic push``. In this case, GCC pretends that there is a ``#pragma diagnostic push`` at the very beginning of the source file, so "unpaired" ``#pragma diagnostic pop`` matches that implicit push. This makes a difference for ``#pragma GCC diagnostic ignored`` @@ -1406,7 +1406,7 @@ project even if there are violations in some headers. # directory. But it'll still complain for all the other sources, e.g: $ cat foo/bar.cc #include "dir/include.h" // Clang flags unused declarations here. - #include "foo/include.h" // but unused warnings under this source is omitted. + #include "foo/include.h" // but unused warnings under this source are omitted. #include "next_to_bar_cc.h" // as are unused warnings from this header file. // Further, unused warnings in the remainder of bar.cc are also omitted. @@ -1648,7 +1648,7 @@ for more details. .. option:: -fno-fast-math - Disable fast-math mode. This options disables unsafe floating-point + Disable fast-math mode. This option disables unsafe floating-point optimizations by preventing the compiler from making any transformations that could affect the results. @@ -1766,7 +1766,7 @@ for more details. * ``fast``: enable fusion across statements disregarding pragmas, breaking compliance with the C and C++ standards (default for CUDA). - * ``on``: enable C and C++ standard complaint fusion in the same statement + * ``on``: enable C and C++ standard compliant fusion in the same statement unless dictated by pragmas (default for languages other than CUDA/HIP) * ``off``: disable fusion * ``fast-honor-pragmas``: fuse across statements unless dictated by pragmas @@ -1919,7 +1919,7 @@ for more details. a single expression of the code. Valid values are: ``source``, ``double``, and ``extended``. - For 64-bit targets, the default value is ``source``. For 32-bit x86 targets + For 64-bit targets, the default value is ``source``. For 32-bit x86 targets, however, in the case of NETBSD 6.99.26 and under, the default value is ``double``; in the case of NETBSD greater than 6.99.26, with NoSSE, the default value is ``extended``, with SSE the default value is ``source``. @@ -3881,9 +3881,9 @@ See :doc:`LanguageExtensions`. Differences between various standard modes ------------------------------------------ -clang supports the -std option, which changes what language mode clang uses. +clang supports the ``-std`` option, which changes what language mode clang uses. The supported modes for C are c89, gnu89, c94, c99, gnu99, c11, gnu11, c17, -gnu17, c23, gnu23, c2y, gnu2y, and various aliases for those modes. If no -std +gnu17, c23, gnu23, c2y, gnu2y, and various aliases for those modes. If no ``-std`` option is specified, clang defaults to gnu17 mode. Many C99 and C11 features are supported in earlier modes as a conforming extension, with a warning. Use ``-pedantic-errors`` to request an error if a feature from a later standard @@ -4609,7 +4609,7 @@ codebases. On ``x86_64-mingw32``, passing i128(by value) is incompatible with the Microsoft x64 calling convention. You might need to tweak -``WinX86_64ABIInfo::classify()`` in lib/CodeGen/Targets/X86.cpp. +``WinX86_64ABIInfo::classify()`` in ``lib/CodeGen/Targets/X86.cpp``. For the X86 target, clang supports the `-m16` command line argument which enables 16-bit code output. This is broadly similar to @@ -4760,8 +4760,8 @@ is imported, the linker will generate fixup code for reading or writing to the variable. When multiple toc-data options are used, the last option used has the affect. -For example: -mno-tocdata=g5,g1 -mtocdata=g1,g2 -mno-tocdata=g2 -mtocdata=g3,g4 -results in -mtocdata=g1,g3,g4 +For example: ``-mno-tocdata=g5,g1 -mtocdata=g1,g2 -mno-tocdata=g2 -mtocdata=g3,g4`` +results in ``-mtocdata=g1,g3,g4`` Names of variables not having external linkage will be ignored. @@ -5143,16 +5143,16 @@ Execute ``clang-cl /?`` to see a list of supported options: Instrument only functions from files where names match any regex separated by a semi-colon -fprofile-generate=<dirname> Generate instrumented code to collect execution counts into a raw profile file in the directory specified by the argument. The filename uses default_%m.profraw pattern - (overridden by LLVM_PROFILE_FILE env var) + (overridden by ``LLVM_PROFILE_FILE`` env var) -fprofile-generate Generate instrumented code to collect execution counts into default_%m.profraw file - (overridden by '=' form of option or LLVM_PROFILE_FILE env var) + (overridden by '=' form of option or ``LLVM_PROFILE_FILE`` env var) -fprofile-instr-generate=<file_name_pattern> Generate instrumented code to collect execution counts into the file whose name pattern is specified as the argument - (overridden by LLVM_PROFILE_FILE env var) + (overridden by ``LLVM_PROFILE_FILE`` env var) -fprofile-instr-generate Generate instrumented code to collect execution counts into default.profraw file - (overridden by '=' form of option or LLVM_PROFILE_FILE env var) + (overridden by '=' form of option or ``LLVM_PROFILE_FILE`` env var) -fprofile-instr-use=<value> Use instrumentation data for coverage testing or profile-guided optimization -fprofile-use=<value> diff --git a/clang/include/clang/Basic/BuiltinsHexagon.td b/clang/include/clang/Basic/BuiltinsHexagon.td index 0727c67..cf18359 100644 --- a/clang/include/clang/Basic/BuiltinsHexagon.td +++ b/clang/include/clang/Basic/BuiltinsHexagon.td @@ -20,8 +20,9 @@ class V<string version, VFeatures newer> : VFeatures { let Features = !strconcat("v", version, "|", newer.Features); } -let Features = "v79" in def V79 : VFeatures; +let Features = "v81" in def V81 : VFeatures; +def V79 : V<"79", V81>; def V75 : V<"75", V79>; def V73 : V<"73", V75>; def V71 : V<"71", V73>; @@ -43,8 +44,9 @@ class HVXV<string version, HVXVFeatures newer> : HVXVFeatures { let Features = !strconcat("hvxv", version, "|", newer.Features); } -let Features = "hvxv79" in def HVXV79 : HVXVFeatures; +let Features = "hvxv81" in def HVXV81 : HVXVFeatures; +def HVXV79 : HVXV<"79", HVXV81>; def HVXV75 : HVXV<"75", HVXV79>; def HVXV73 : HVXV<"73", HVXV75>; def HVXV71 : HVXV<"71", HVXV73>; diff --git a/clang/include/clang/Basic/BuiltinsX86.td b/clang/include/clang/Basic/BuiltinsX86.td index 8332eac..0c85e28 100644 --- a/clang/include/clang/Basic/BuiltinsX86.td +++ b/clang/include/clang/Basic/BuiltinsX86.td @@ -198,6 +198,10 @@ let Features = "sse", Header = "xmmintrin.h", Attributes = [NoThrow, RequireDecl def _mm_sfence : X86LibBuiltin<"void()">; } +let Features = "sse", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in { + def shufps : X86Builtin<"_Vector<4, float>(_Vector<4, float>, _Vector<4, float>, _Constant int)">; +} + let Features = "sse", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in { def rcpps : X86Builtin<"_Vector<4, float>(_Vector<4, float>)">; def rcpss : X86Builtin<"_Vector<4, float>(_Vector<4, float>)">; @@ -205,7 +209,6 @@ let Features = "sse", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in def rsqrtss : X86Builtin<"_Vector<4, float>(_Vector<4, float>)">; def sqrtps : X86Builtin<"_Vector<4, float>(_Vector<4, float>)">; def sqrtss : X86Builtin<"_Vector<4, float>(_Vector<4, float>)">; - def shufps : X86Builtin<"_Vector<4, float>(_Vector<4, float>, _Vector<4, float>, _Constant int)">; } let Features = "sse2", Attributes = [NoThrow, RequiredVectorWidth<128>] in { @@ -222,13 +225,13 @@ let Features = "sse2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWi def pshufhw : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Constant int)">; def movmskpd : X86Builtin<"int(_Vector<2, double>)">; def pmovmskb128 : X86Builtin<"int(_Vector<16, char>)">; + def shufpd : X86Builtin<"_Vector<2, double>(_Vector<2, double>, _Vector<2, double>, _Constant int)">; } let Features = "sse2", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in { def psadbw128 : X86Builtin<"_Vector<2, long long int>(_Vector<16, char>, _Vector<16, char>)">; def sqrtpd : X86Builtin<"_Vector<2, double>(_Vector<2, double>)">; def sqrtsd : X86Builtin<"_Vector<2, double>(_Vector<2, double>)">; - def shufpd : X86Builtin<"_Vector<2, double>(_Vector<2, double>, _Vector<2, double>, _Constant int)">; def cvtpd2dq : X86Builtin<"_Vector<2, long long int>(_Vector<2, double>)">; def cvtpd2ps : X86Builtin<"_Vector<4, float>(_Vector<2, double>)">; def cvttpd2dq : X86Builtin<"_Vector<4, int>(_Vector<2, double>)">; @@ -280,8 +283,6 @@ let Features = "sse2", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] i def psllw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">; def pslld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">; def psllq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">; - def pslldqi128_byteshift : X86Builtin<"_Vector<16, char>(_Vector<16, char>, _Constant int)">; - def psrldqi128_byteshift : X86Builtin<"_Vector<16, char>(_Vector<16, char>, _Constant int)">; } let Features = "sse2", @@ -300,6 +301,9 @@ let Features = "sse2", def psrawi128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, int)">; def psradi128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, int)">; + + def pslldqi128_byteshift : X86Builtin<"_Vector<16, char>(_Vector<16, char>, _Constant int)">; + def psrldqi128_byteshift : X86Builtin<"_Vector<16, char>(_Vector<16, char>, _Constant int)">; } let Features = "sse3", Attributes = [NoThrow] in { @@ -487,13 +491,16 @@ let Features = "avx512f,vpclmulqdq", Attributes = [NoThrow, Const, RequiredVecto def pclmulqdq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Vector<8, long long int>, _Constant char)">; } +let Features = "avx", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in { + def shufpd256 : X86Builtin<"_Vector<4, double>(_Vector<4, double>, _Vector<4, double>, _Constant int)">; + def shufps256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, float>, _Constant int)">; +} + let Features = "avx", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in { def vpermilvarpd : X86Builtin<"_Vector<2, double>(_Vector<2, double>, _Vector<2, long long int>)">; def vpermilvarps : X86Builtin<"_Vector<4, float>(_Vector<4, float>, _Vector<4, int>)">; def vpermilvarpd256 : X86Builtin<"_Vector<4, double>(_Vector<4, double>, _Vector<4, long long int>)">; def vpermilvarps256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, int>)">; - def shufpd256 : X86Builtin<"_Vector<4, double>(_Vector<4, double>, _Vector<4, double>, _Constant int)">; - def shufps256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, float>, _Constant int)">; def dpps256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, float>, _Constant char)">; def cmppd256 : X86Builtin<"_Vector<4, double>(_Vector<4, double>, _Vector<4, double>, _Constant char)">; def cmpps256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, float>, _Constant char)">; @@ -607,12 +614,10 @@ let Features = "avx2", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] i : X86Builtin< "_Vector<4, long long int>(_Vector<32, char>, _Vector<32, char>)">; def psllw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">; - def pslldqi256_byteshift : X86Builtin<"_Vector<32, char>(_Vector<32, char>, _Constant int)">; def pslld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">; def psllq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">; def psraw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">; def psrad256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">; - def psrldqi256_byteshift : X86Builtin<"_Vector<32, char>(_Vector<32, char>, _Constant int)">; def psrlw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">; def psrld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">; def psrlq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">; @@ -646,10 +651,12 @@ let Features = "avx2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWi def psllwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">; def pslldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">; def psllqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">; + def pslldqi256_byteshift : X86Builtin<"_Vector<32, char>(_Vector<32, char>, _Constant int)">; def psrlwi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">; def psrldi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">; def psrlqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">; + def psrldqi256_byteshift : X86Builtin<"_Vector<32, char>(_Vector<32, char>, _Constant int)">; def psrawi256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, int)">; def psradi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, int)">; @@ -2090,6 +2097,9 @@ let Features = "avx512bw", Attributes = [NoThrow, Const, RequiredVectorWidth<512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">; def psrlw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">; +} + +let Features = "avx512bw", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in { def pslldqi512_byteshift : X86Builtin<"_Vector<64, char>(_Vector<64, char>, _Constant int)">; def psrldqi512_byteshift : X86Builtin<"_Vector<64, char>(_Vector<64, char>, _Constant int)">; } @@ -2470,6 +2480,9 @@ let Features = "avx512f", Attributes = [NoThrow, Const, RequiredVectorWidth<512> def shuf_f64x2 : X86Builtin<"_Vector<8, double>(_Vector<8, double>, _Vector<8, double>, _Constant int)">; def shuf_i32x4 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, _Vector<16, int>, _Constant int)">; def shuf_i64x2 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Vector<8, long long int>, _Constant int)">; +} + +let Features = "avx512f", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in { def shufpd512 : X86Builtin<"_Vector<8, double>(_Vector<8, double>, _Vector<8, double>, _Constant int)">; def shufps512 : X86Builtin<"_Vector<16, float>(_Vector<16, float>, _Vector<16, float>, _Constant int)">; } diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td index 13f0d59..6087893 100644 --- a/clang/include/clang/Basic/DiagnosticSemaKinds.td +++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td @@ -2418,9 +2418,9 @@ def err_init_conversion_failed : Error< "cannot initialize %select{a variable|a parameter|template parameter|" "return object|statement expression result|an " "exception object|a member subobject|an array element|a new value|a value|a " - "base class|a constructor delegation|a vector element|a block element|a " - "block element|a complex element|a lambda capture|a compound literal " - "initializer|a related result|a parameter of CF audited function|a " + "base class|a constructor delegation|a vector element|a matrix element|a " + "block element|a block element|a complex element|a lambda capture|a compound" + " literal initializer|a related result|a parameter of CF audited function|a " "structured binding|a member subobject}0 " "%diff{of type $ with an %select{rvalue|lvalue}2 of type $|" "with an %select{rvalue|lvalue}2 of incompatible type}1,3" @@ -6549,9 +6549,9 @@ def warn_extern_init : Warning<"'extern' variable has an initializer">, def err_variable_object_no_init : Error< "variable-sized object may not be initialized">; def err_excess_initializers : Error< - "excess elements in %select{array|vector|scalar|union|struct}0 initializer">; + "excess elements in %select{array|vector|matrix|scalar|union|struct}0 initializer">; def ext_excess_initializers : ExtWarn< - "excess elements in %select{array|vector|scalar|union|struct}0 initializer">, + "excess elements in %select{array|vector|matrix|scalar|union|struct}0 initializer">, InGroup<ExcessInitializers>; def err_excess_initializers_for_sizeless_type : Error< "excess elements in initializer for indivisible sizeless type %0">; @@ -13148,6 +13148,7 @@ def err_riscv_attribute_interrupt_requires_extension : Error< "RISC-V 'interrupt' attribute '%0' requires extension '%1'">; def err_riscv_attribute_interrupt_invalid_combination : Error< "RISC-V 'interrupt' attribute contains invalid combination of interrupt types">; +def err_riscv_builtin_invalid_twiden : Error<"RISC-V XSfmm twiden must be 1, 2 or 4">; def err_std_source_location_impl_not_found : Error< "'std::source_location::__impl' was not found; it must be defined before '__builtin_source_location' is called">; diff --git a/clang/include/clang/Basic/FileManager.h b/clang/include/clang/Basic/FileManager.h index 337911e..fa7552b 100644 --- a/clang/include/clang/Basic/FileManager.h +++ b/clang/include/clang/Basic/FileManager.h @@ -287,8 +287,12 @@ public: /// If path is not absolute and FileSystemOptions set the working /// directory, the path is modified to be relative to the given /// working directory. - /// \returns true if \c path changed. - bool FixupRelativePath(SmallVectorImpl<char> &path) const; + /// \returns true if \c Path changed. + bool FixupRelativePath(SmallVectorImpl<char> &Path) const { + return fixupRelativePath(FileSystemOpts, Path); + } + static bool fixupRelativePath(const FileSystemOptions &FileSystemOpts, + SmallVectorImpl<char> &Path); /// Makes \c Path absolute taking into account FileSystemOptions and the /// working directory option. diff --git a/clang/include/clang/Basic/riscv_sifive_vector.td b/clang/include/clang/Basic/riscv_sifive_vector.td index 772fd3e..89e644a 100644 --- a/clang/include/clang/Basic/riscv_sifive_vector.td +++ b/clang/include/clang/Basic/riscv_sifive_vector.td @@ -14,6 +14,10 @@ include "riscv_vector_common.td" +class IsFloat<string type> { + bit val = !or(!eq(type, "x"), !eq(type, "f"), !eq(type, "d"), !eq(type, "y")); +} + //===----------------------------------------------------------------------===// // Instruction definitions //===----------------------------------------------------------------------===// @@ -198,3 +202,169 @@ let ManualCodegen = [{ defm sf_vfnrclip_xu_f_qf : RVVVFNRCLIPBuiltinSet<"Uv", "UvFqf", "c">; } } + +multiclass RVVSFTileLoadStoreBuiltinSet<list<string> types, + list<string> RequiredFeatures = []> { + let OverloadedName = NAME, + Name = NAME, + IRName = NAME, + Log2LMUL = [0], + HasMasked = false, + ManualCodegen = [{IntrinsicTypes = {Ops.back()->getType()};}] in + foreach type = types in { + let RequiredFeatures = !listconcat(RequiredFeatures, + !cond(!eq(type, "x"): ["zvfhmin"], + !eq(type, "y"): ["zvfbfmin"], + true: []<string>)) in { + def : RVVBuiltin<"e", "0zPCe", type>; + if !not(IsFloat<type>.val) then + def : RVVBuiltin<"Ue", "0zPCUe", type>; + } + } +} + +multiclass RVVSFTileMoveBuiltinSet<list<list<string>> suffixes_prototypes, + list<int> intrinsic_types, + string type, + list<string> RequiredFeatures = []> { + foreach sp = suffixes_prototypes in + let RequiredFeatures = !listconcat(RequiredFeatures, + !cond(!eq(type, "x"): ["zvfhmin"], + !eq(type, "y"): ["zvfbfmin"], + true: []<string>)), + SupportOverloading = false, + HasMasked = false, + Name = NAME, + IRName = NAME, + HasVL = true, + Log2LMUL = [3], + IntrinsicTypes = intrinsic_types in + def : RVVBuiltin<sp[0], sp[1], type>; +} + +multiclass RVVSFTileMoveVTBuiltinSet<list<string> RequiredFeatures = []> { + foreach type = ["c", "s", "i", "l"] in + defm NAME : + RVVSFTileMoveBuiltinSet<[["v", "vz"], ["Uv", "Uvz"]], [-1], type, + RequiredFeatures>; + foreach type = ["x", "y", "f", "d"] in + defm NAME : + RVVSFTileMoveBuiltinSet<[["v", "vz"]], [-1], type, RequiredFeatures>; +} + +multiclass RVVSFTileMoveTVBuiltinSet<list<string> RequiredFeatures = []> { + let SupportOverloading = true, OverloadedName = NAME in { + foreach type = ["c", "s", "i", "l"] in + defm NAME : + RVVSFTileMoveBuiltinSet<[["v", "0zv"], ["Uv", "0zUv"]], [1], type, + RequiredFeatures>; + foreach type = ["x", "y", "f", "d"] in + defm NAME : + RVVSFTileMoveBuiltinSet<[["v", "0zv"]], [1], type, RequiredFeatures>; + } +} + +multiclass RVVOp0Op1Op2BuiltinSet<string intrinsic_name, string type_range, + list<list<string>> suffixes_prototypes> + : RVVBuiltinSet<intrinsic_name, type_range, suffixes_prototypes, [0, 1, 2]>; + +multiclass RVVSFMatMulBuiltinSet<string prototype, string suffix, + string type_range, list<int> widens> { + foreach widen = widens in + let OverloadedName = NAME, + TWiden = widen, + HasVL = false, + Log2LMUL = [3], + HasMasked = false in + defm NAME : RVVOp0Op1Op2BuiltinSet<NAME, type_range, + [[!strconcat("w", !cast<string>(widen)), suffix, prototype]]>; +} + +multiclass RVVSFMatMulFloatBuiltinSet<string name, string prototype, string suffix, + list<string> type_range, int widen> { + // Currently the XSfmm spec doesn't support w8. + foreach type = type_range in + let OverloadedName = name # !strconcat("_w", !cast<string>(widen)), + TWiden = widen, + HasVL = false, + Log2LMUL = [3], + Name = name # "_" # !strconcat("w", !cast<string>(widen)), + HasMasked = false in + defm : RVVOp0Op1BuiltinSet<name, type, [["", suffix, prototype]]>; +} + +multiclass RVVSFVTZeroBuiltinSet { + let SupportOverloading = false, + HasVL = false, + HasMasked = false, + Name = NAME, + IRName = NAME, + Log2LMUL = [0] in + defm : RVVOp0BuiltinSet<NAME, "i", [["", "", "0Kzzzzz"]]>; +} + +multiclass RVVSFVTDiscardBuiltinSet { + let SupportOverloading = false, + HasVL = false, + HasMasked = false, + Name = NAME, + IRName = NAME, + Log2LMUL = [0] in + defm : RVVBuiltinSet<NAME, "i", [["", "", "0"]], []>; +} + +let RequiredFeatures = ["xsfmmbase"] in { + let SupportOverloading = false, + HasVL = false, + HasMasked = false, + Log2LMUL = [0], + ManualCodegen = [{IntrinsicTypes = {ResultType};}] in // Set XLEN type + { + // let HasBuiltinAlias = false in + def sf_vsettnt : RVVBuiltin<"", "zzKzKz", "i">; + def sf_vsettm : RVVBuiltin<"", "zzKzKz", "i">; + let IRName = "sf_vsettnt" in + def sf_vsettn : RVVBuiltin<"", "zzKzKz", "i">; + def sf_vsettk : RVVBuiltin<"", "zzKzKz", "i">; + } + defm sf_vtzero_t : RVVSFVTZeroBuiltinSet; + defm sf_vtdiscard : RVVSFVTDiscardBuiltinSet; +} + +defm sf_vtmv_v_t : RVVSFTileMoveVTBuiltinSet<["xsfmmbase"]>; +defm sf_vtmv_t_v : RVVSFTileMoveTVBuiltinSet<["xsfmmbase"]>; + +defm sf_vlte8 : RVVSFTileLoadStoreBuiltinSet<["c"], ["xsfmmbase"]>; +defm sf_vlte16 : RVVSFTileLoadStoreBuiltinSet<["s", "x", "y"], ["xsfmmbase"]>; +defm sf_vlte32 : RVVSFTileLoadStoreBuiltinSet<["i", "f"], ["xsfmmbase"]>; +defm sf_vlte64 : RVVSFTileLoadStoreBuiltinSet<["l", "d"], ["xsfmmbase"]>; + +defm sf_vste8 : RVVSFTileLoadStoreBuiltinSet<["c"], ["xsfmmbase"]>; +defm sf_vste16 : RVVSFTileLoadStoreBuiltinSet<["s", "x", "y"], ["xsfmmbase"]>; +defm sf_vste32 : RVVSFTileLoadStoreBuiltinSet<["i", "f"], ["xsfmmbase"]>; +defm sf_vste64 : RVVSFTileLoadStoreBuiltinSet<["l", "d"], ["xsfmmbase"]>; + +let RequiredFeatures = ["xsfmm32a8i"] in { + defm sf_mm_u_u : RVVSFMatMulBuiltinSet<"0KzUvUvzzz", "UvUv", "c", [4]>; + defm sf_mm_s_u : RVVSFMatMulBuiltinSet<"0KzvUvzzz", "vUv", "c", [4]>; + defm sf_mm_u_s : RVVSFMatMulBuiltinSet<"0KzUvvzzz", "Uvv", "c", [4]>; + defm sf_mm_s_s : RVVSFMatMulBuiltinSet<"0Kzvvzzz", "vv", "c", [4]>; + +} + +let RequiredFeatures = ["xsfmm32a16f"] in + defm : RVVSFMatMulFloatBuiltinSet<"sf_mm_f_f", "0Kzvvzzz", "v", ["x", "y"], 2>; + +let RequiredFeatures = ["xsfmm32a32f"] in + defm : RVVSFMatMulFloatBuiltinSet<"sf_mm_f_f", "0Kzvvzzz", "v", ["f"], 1>; + +let RequiredFeatures = ["xsfmm32a8f"] in + foreach e1 = [5, 4] in + foreach e2 = [5, 4] in + let OverloadedName = "sf_mm_e" # e1 # "m" # !sub(7, e1) # "_e" # e2 # "m" # !sub(7, e2) in + defm : RVVSFMatMulFloatBuiltinSet< + "sf_mm_e" # e1 # "m" # !sub(7, e1) # "_e" # e2 # "m" # !sub(7, e2), + "0KzUvUvzzz", "UvUv", ["c"], 4>; + +let RequiredFeatures = ["xsfmm64a64f"] in + defm : RVVSFMatMulFloatBuiltinSet<"sf_mm_f_f", "0Kzvvzzz", "v", ["d"], 1>; diff --git a/clang/include/clang/Basic/riscv_vector_common.td b/clang/include/clang/Basic/riscv_vector_common.td index eaa2ba4..7e2d339 100644 --- a/clang/include/clang/Basic/riscv_vector_common.td +++ b/clang/include/clang/Basic/riscv_vector_common.td @@ -247,6 +247,9 @@ class RVVBuiltin<string suffix, string prototype, string type_range, // Set to true if the builtin has a parameter that models floating-point // rounding mode control bit HasFRMRoundModeOp = false; + + // TWiden for XSfmm. + int TWiden = 0; } // This is the code emitted in the header. diff --git a/clang/include/clang/CIR/MissingFeatures.h b/clang/include/clang/CIR/MissingFeatures.h index 598e826a..48ef8be 100644 --- a/clang/include/clang/CIR/MissingFeatures.h +++ b/clang/include/clang/CIR/MissingFeatures.h @@ -150,11 +150,9 @@ struct MissingFeatures { static bool zeroSizeRecordMembers() { return false; } // Coroutines - static bool coroAllocBuiltinCall() { return false; } - static bool coroBeginBuiltinCall() { return false; } static bool coroEndBuiltinCall() { return false; } - static bool coroSizeBuiltinCall() { return false; } static bool coroutineFrame() { return false; } + static bool emitBodyAndFallthrough() { return false; } // Various handling of deferred processing in CIRGenModule. static bool cgmRelease() { return false; } @@ -231,6 +229,7 @@ struct MissingFeatures { static bool coverageMapping() { return false; } static bool createInvariantGroup() { return false; } static bool createProfileWeightsForLoop() { return false; } + static bool ctorConstLvalueToRvalueConversion() { return false; } static bool ctorMemcpyizer() { return false; } static bool cudaSupport() { return false; } static bool cxxRecordStaticMembers() { return false; } @@ -240,6 +239,7 @@ struct MissingFeatures { static bool dataLayoutPtrHandlingBasedOnLangAS() { return false; } static bool deferredCXXGlobalInit() { return false; } static bool deleteArray() { return false; } + static bool devirtualizeDestructor() { return false; } static bool devirtualizeMemberFunction() { return false; } static bool ehCleanupFlags() { return false; } static bool ehCleanupHasPrebranchedFallthrough() { return false; } diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td index 0c9584f..93aeb22 100644 --- a/clang/include/clang/Driver/Options.td +++ b/clang/include/clang/Driver/Options.td @@ -6558,6 +6558,8 @@ def mv75 : Flag<["-"], "mv75">, Group<m_hexagon_Features_Group>, Alias<mcpu_EQ>, AliasArgs<["hexagonv75"]>; def mv79 : Flag<["-"], "mv79">, Group<m_hexagon_Features_Group>, Alias<mcpu_EQ>, AliasArgs<["hexagonv79"]>; +def mv81 : Flag<["-"], "mv81">, Group<m_hexagon_Features_Group>, + Alias<mcpu_EQ>, AliasArgs<["hexagonv81"]>; def mhexagon_hvx : Flag<["-"], "mhvx">, Group<m_hexagon_Features_HVX_Group>, HelpText<"Enable Hexagon Vector eXtensions">; def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">, diff --git a/clang/include/clang/Sema/Initialization.h b/clang/include/clang/Sema/Initialization.h index 5e96317..0e2891f 100644 --- a/clang/include/clang/Sema/Initialization.h +++ b/clang/include/clang/Sema/Initialization.h @@ -91,6 +91,10 @@ public: /// or vector. EK_VectorElement, + /// The entity being initialized is an element of a matrix. + /// or matrix. + EK_MatrixElement, + /// The entity being initialized is a field of block descriptor for /// the copied-in c++ object. EK_BlockElement, @@ -205,8 +209,8 @@ private: /// virtual base. llvm::PointerIntPair<const CXXBaseSpecifier *, 1> Base; - /// When Kind == EK_ArrayElement, EK_VectorElement, or - /// EK_ComplexElement, the index of the array or vector element being + /// When Kind == EK_ArrayElement, EK_VectorElement, EK_MatrixElement, + /// or EK_ComplexElement, the index of the array or vector element being /// initialized. unsigned Index; @@ -536,7 +540,7 @@ public: /// element's index. unsigned getElementIndex() const { assert(getKind() == EK_ArrayElement || getKind() == EK_VectorElement || - getKind() == EK_ComplexElement); + getKind() == EK_MatrixElement || getKind() == EK_ComplexElement); return Index; } @@ -544,7 +548,7 @@ public: /// element, sets the element index. void setElementIndex(unsigned Index) { assert(getKind() == EK_ArrayElement || getKind() == EK_VectorElement || - getKind() == EK_ComplexElement); + getKind() == EK_MatrixElement || getKind() == EK_ComplexElement); this->Index = Index; } diff --git a/clang/include/clang/Sema/Sema.h b/clang/include/clang/Sema/Sema.h index 189798f..52904c72 100644 --- a/clang/include/clang/Sema/Sema.h +++ b/clang/include/clang/Sema/Sema.h @@ -2681,11 +2681,6 @@ public: /// function without this attribute. bool DiscardingCFIUncheckedCallee(QualType From, QualType To) const; - /// Returns true if `From` is a function or pointer to a function without the - /// `cfi_unchecked_callee` attribute but `To` is a function or pointer to - /// function with this attribute. - bool AddingCFIUncheckedCallee(QualType From, QualType To) const; - /// This function calls Action when it determines that E designates a /// misaligned member due to the packed attribute. This is used to emit /// local diagnostics like in reference binding. diff --git a/clang/include/clang/Support/RISCVVIntrinsicUtils.h b/clang/include/clang/Support/RISCVVIntrinsicUtils.h index 29a07f1..67149f2 100644 --- a/clang/include/clang/Support/RISCVVIntrinsicUtils.h +++ b/clang/include/clang/Support/RISCVVIntrinsicUtils.h @@ -402,6 +402,7 @@ private: std::vector<int64_t> IntrinsicTypes; unsigned NF = 1; Policy PolicyAttrs; + unsigned TWiden = 0; public: RVVIntrinsic(llvm::StringRef Name, llvm::StringRef Suffix, @@ -410,8 +411,8 @@ public: bool HasVL, PolicyScheme Scheme, bool SupportOverloading, bool HasBuiltinAlias, llvm::StringRef ManualCodegen, const RVVTypes &Types, - const std::vector<int64_t> &IntrinsicTypes, - unsigned NF, Policy PolicyAttrs, bool HasFRMRoundModeOp); + const std::vector<int64_t> &IntrinsicTypes, unsigned NF, + Policy PolicyAttrs, bool HasFRMRoundModeOp, unsigned TWiden); ~RVVIntrinsic() = default; RVVTypePtr getOutputType() const { return OutputType; } @@ -435,6 +436,7 @@ public: llvm::StringRef getManualCodegen() const { return ManualCodegen; } PolicyScheme getPolicyScheme() const { return Scheme; } unsigned getNF() const { return NF; } + unsigned getTWiden() const { return TWiden; } const std::vector<int64_t> &getIntrinsicTypes() const { return IntrinsicTypes; } diff --git a/clang/lib/AST/ByteCode/Compiler.cpp b/clang/lib/AST/ByteCode/Compiler.cpp index 836d22f..f4ddbf4 100644 --- a/clang/lib/AST/ByteCode/Compiler.cpp +++ b/clang/lib/AST/ByteCode/Compiler.cpp @@ -3273,34 +3273,43 @@ bool Compiler<Emitter>::VisitCXXConstructExpr(const CXXConstructExpr *E) { } if (T->isArrayType()) { - const ConstantArrayType *CAT = - Ctx.getASTContext().getAsConstantArrayType(E->getType()); - if (!CAT) - return false; - - size_t NumElems = CAT->getZExtSize(); const Function *Func = getFunction(E->getConstructor()); if (!Func) return false; - // FIXME(perf): We're calling the constructor once per array element here, - // in the old intepreter we had a special-case for trivial constructors. - for (size_t I = 0; I != NumElems; ++I) { - if (!this->emitConstUint64(I, E)) - return false; - if (!this->emitArrayElemPtrUint64(E)) - return false; + if (!this->emitDupPtr(E)) + return false; - // Constructor arguments. - for (const auto *Arg : E->arguments()) { - if (!this->visit(Arg)) - return false; + std::function<bool(QualType)> initArrayDimension; + initArrayDimension = [&](QualType T) -> bool { + if (!T->isArrayType()) { + // Constructor arguments. + for (const auto *Arg : E->arguments()) { + if (!this->visit(Arg)) + return false; + } + + return this->emitCall(Func, 0, E); } - if (!this->emitCall(Func, 0, E)) + const ConstantArrayType *CAT = + Ctx.getASTContext().getAsConstantArrayType(T); + if (!CAT) return false; - } - return true; + QualType ElemTy = CAT->getElementType(); + unsigned NumElems = CAT->getZExtSize(); + for (size_t I = 0; I != NumElems; ++I) { + if (!this->emitConstUint64(I, E)) + return false; + if (!this->emitArrayElemPtrUint64(E)) + return false; + if (!initArrayDimension(ElemTy)) + return false; + } + return this->emitPopPtr(E); + }; + + return initArrayDimension(E->getType()); } return false; @@ -3599,8 +3608,6 @@ bool Compiler<Emitter>::VisitCXXNewExpr(const CXXNewExpr *E) { if (PlacementDest) { if (!this->visit(PlacementDest)) return false; - if (!this->emitStartLifetime(E)) - return false; if (!this->emitGetLocal(SizeT, ArrayLen, E)) return false; if (!this->emitCheckNewTypeMismatchArray(SizeT, E, E)) @@ -3740,10 +3747,9 @@ bool Compiler<Emitter>::VisitCXXNewExpr(const CXXNewExpr *E) { if (PlacementDest) { if (!this->visit(PlacementDest)) return false; - if (!this->emitStartLifetime(E)) - return false; if (!this->emitCheckNewTypeMismatch(E, E)) return false; + } else { // Allocate just one element. if (!this->emitAlloc(Desc, E)) diff --git a/clang/lib/AST/ByteCode/Interp.cpp b/clang/lib/AST/ByteCode/Interp.cpp index a72282c..169a9a2 100644 --- a/clang/lib/AST/ByteCode/Interp.cpp +++ b/clang/lib/AST/ByteCode/Interp.cpp @@ -1903,12 +1903,19 @@ bool CheckNewTypeMismatch(InterpState &S, CodePtr OpPC, const Expr *E, if (Ptr.inUnion() && Ptr.getBase().getRecord()->isUnion()) Ptr.activate(); + if (Ptr.isZero()) { + S.FFDiag(S.Current->getSource(OpPC), diag::note_constexpr_access_null) + << AK_Construct; + return false; + } + if (!Ptr.isBlockPointer()) return false; + startLifetimeRecurse(Ptr); + // Similar to CheckStore(), but with the additional CheckTemporary() call and // the AccessKinds are different. - if (!Ptr.block()->isAccessible()) { if (!CheckExtern(S, OpPC, Ptr)) return false; diff --git a/clang/lib/AST/ByteCode/InterpBuiltin.cpp b/clang/lib/AST/ByteCode/InterpBuiltin.cpp index ff50e6d..d0b97a1 100644 --- a/clang/lib/AST/ByteCode/InterpBuiltin.cpp +++ b/clang/lib/AST/ByteCode/InterpBuiltin.cpp @@ -3320,6 +3320,65 @@ static bool interp__builtin_ia32_vpconflict(InterpState &S, CodePtr OpPC, return true; } +static bool interp__builtin_x86_byteshift( + InterpState &S, CodePtr OpPC, const CallExpr *Call, unsigned ID, + llvm::function_ref<APInt(const Pointer &, unsigned Lane, unsigned I, + unsigned Shift)> + Fn) { + assert(Call->getNumArgs() == 2); + + APSInt ImmAPS = popToAPSInt(S, Call->getArg(1)); + uint64_t Shift = ImmAPS.getZExtValue() & 0xff; + + const Pointer &Src = S.Stk.pop<Pointer>(); + if (!Src.getFieldDesc()->isPrimitiveArray()) + return false; + + unsigned NumElems = Src.getNumElems(); + const Pointer &Dst = S.Stk.peek<Pointer>(); + PrimType ElemT = Src.getFieldDesc()->getPrimType(); + + for (unsigned Lane = 0; Lane != NumElems; Lane += 16) { + for (unsigned I = 0; I != 16; ++I) { + unsigned Base = Lane + I; + APSInt Result = APSInt(Fn(Src, Lane, I, Shift)); + INT_TYPE_SWITCH_NO_BOOL(ElemT, + { Dst.elem<T>(Base) = static_cast<T>(Result); }); + } + } + + Dst.initializeAllElements(); + + return true; +} + +static bool interp__builtin_ia32_shuffle_generic( + InterpState &S, CodePtr OpPC, const CallExpr *Call, + llvm::function_ref<std::pair<unsigned, unsigned>(unsigned, unsigned)> + GetSourceIndex) { + + assert(Call->getNumArgs() == 3); + unsigned ShuffleMask = popToAPSInt(S, Call->getArg(2)).getZExtValue(); + + QualType Arg0Type = Call->getArg(0)->getType(); + const auto *VecT = Arg0Type->castAs<VectorType>(); + PrimType ElemT = *S.getContext().classify(VecT->getElementType()); + unsigned NumElems = VecT->getNumElements(); + + const Pointer &B = S.Stk.pop<Pointer>(); + const Pointer &A = S.Stk.pop<Pointer>(); + const Pointer &Dst = S.Stk.peek<Pointer>(); + + for (unsigned DstIdx = 0; DstIdx != NumElems; ++DstIdx) { + auto [SrcVecIdx, SrcIdx] = GetSourceIndex(DstIdx, ShuffleMask); + const Pointer &Src = (SrcVecIdx == 0) ? A : B; + TYPE_SWITCH(ElemT, { Dst.elem<T>(DstIdx) = Src.elem<T>(SrcIdx); }); + } + Dst.initializeAllElements(); + + return true; +} + bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const CallExpr *Call, uint32_t BuiltinID) { if (!S.getASTContext().BuiltinInfo.isConstantEvaluated(BuiltinID)) @@ -4250,6 +4309,42 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const CallExpr *Call, case X86::BI__builtin_ia32_selectpd_512: return interp__builtin_select(S, OpPC, Call); + case X86::BI__builtin_ia32_shufps: + case X86::BI__builtin_ia32_shufps256: + case X86::BI__builtin_ia32_shufps512: + return interp__builtin_ia32_shuffle_generic( + S, OpPC, Call, [](unsigned DstIdx, unsigned ShuffleMask) { + unsigned NumElemPerLane = 4; + unsigned NumSelectableElems = NumElemPerLane / 2; + unsigned BitsPerElem = 2; + unsigned IndexMask = 0x3; + unsigned MaskBits = 8; + unsigned Lane = DstIdx / NumElemPerLane; + unsigned ElemInLane = DstIdx % NumElemPerLane; + unsigned LaneOffset = Lane * NumElemPerLane; + unsigned SrcIdx = ElemInLane >= NumSelectableElems ? 1 : 0; + unsigned BitIndex = (DstIdx * BitsPerElem) % MaskBits; + unsigned Index = (ShuffleMask >> BitIndex) & IndexMask; + return std::pair<unsigned, unsigned>{SrcIdx, LaneOffset + Index}; + }); + case X86::BI__builtin_ia32_shufpd: + case X86::BI__builtin_ia32_shufpd256: + case X86::BI__builtin_ia32_shufpd512: + return interp__builtin_ia32_shuffle_generic( + S, OpPC, Call, [](unsigned DstIdx, unsigned ShuffleMask) { + unsigned NumElemPerLane = 2; + unsigned NumSelectableElems = NumElemPerLane / 2; + unsigned BitsPerElem = 1; + unsigned IndexMask = 0x1; + unsigned MaskBits = 8; + unsigned Lane = DstIdx / NumElemPerLane; + unsigned ElemInLane = DstIdx % NumElemPerLane; + unsigned LaneOffset = Lane * NumElemPerLane; + unsigned SrcIdx = ElemInLane >= NumSelectableElems ? 1 : 0; + unsigned BitIndex = (DstIdx * BitsPerElem) % MaskBits; + unsigned Index = (ShuffleMask >> BitIndex) & IndexMask; + return std::pair<unsigned, unsigned>{SrcIdx, LaneOffset + Index}; + }); case X86::BI__builtin_ia32_pshufb128: case X86::BI__builtin_ia32_pshufb256: case X86::BI__builtin_ia32_pshufb512: @@ -4390,6 +4485,39 @@ bool InterpretBuiltin(InterpState &S, CodePtr OpPC, const CallExpr *Call, case X86::BI__builtin_ia32_vec_set_v4di: return interp__builtin_vec_set(S, OpPC, Call, BuiltinID); + case X86::BI__builtin_ia32_pslldqi128_byteshift: + case X86::BI__builtin_ia32_pslldqi256_byteshift: + case X86::BI__builtin_ia32_pslldqi512_byteshift: + // These SLLDQ intrinsics always operate on byte elements (8 bits). + // The lane width is hardcoded to 16 to match the SIMD register size, + // but the algorithm processes one byte per iteration, + // so APInt(8, ...) is correct and intentional. + return interp__builtin_x86_byteshift( + S, OpPC, Call, BuiltinID, + [](const Pointer &Src, unsigned Lane, unsigned I, unsigned Shift) { + if (I < Shift) { + return APInt(8, 0); + } + return APInt(8, Src.elem<uint8_t>(Lane + I - Shift)); + }); + + case X86::BI__builtin_ia32_psrldqi128_byteshift: + case X86::BI__builtin_ia32_psrldqi256_byteshift: + case X86::BI__builtin_ia32_psrldqi512_byteshift: + // These SRLDQ intrinsics always operate on byte elements (8 bits). + // The lane width is hardcoded to 16 to match the SIMD register size, + // but the algorithm processes one byte per iteration, + // so APInt(8, ...) is correct and intentional. + return interp__builtin_x86_byteshift( + S, OpPC, Call, BuiltinID, + [](const Pointer &Src, unsigned Lane, unsigned I, unsigned Shift) { + if (I + Shift < 16) { + return APInt(8, Src.elem<uint8_t>(Lane + I + Shift)); + } + + return APInt(8, 0); + }); + default: S.FFDiag(S.Current->getLocation(OpPC), diag::note_invalid_subexpr_in_const_expr) diff --git a/clang/lib/AST/ByteCode/Opcodes.td b/clang/lib/AST/ByteCode/Opcodes.td index 406feb5..1c17ad9e 100644 --- a/clang/lib/AST/ByteCode/Opcodes.td +++ b/clang/lib/AST/ByteCode/Opcodes.td @@ -866,19 +866,13 @@ def Free : Opcode { let Args = [ArgBool, ArgBool]; } -def CheckNewTypeMismatch : Opcode { - let Args = [ArgExpr]; -} - -def InvalidNewDeleteExpr : Opcode { - let Args = [ArgExpr]; -} - +def CheckNewTypeMismatch : Opcode { let Args = [ArgExpr]; } def CheckNewTypeMismatchArray : Opcode { let Types = [IntegerTypeClass]; let Args = [ArgExpr]; let HasGroup = 1; } +def InvalidNewDeleteExpr : Opcode { let Args = [ArgExpr]; } def IsConstantContext: Opcode; def CheckAllocations : Opcode; diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp index 2bd4476..29ee089 100644 --- a/clang/lib/AST/ExprConstant.cpp +++ b/clang/lib/AST/ExprConstant.cpp @@ -11619,6 +11619,39 @@ static bool evalPackBuiltin(const CallExpr *E, EvalInfo &Info, APValue &Result, return true; } +static bool evalShuffleGeneric( + EvalInfo &Info, const CallExpr *Call, APValue &Out, + llvm::function_ref<std::pair<unsigned, unsigned>(unsigned, unsigned)> + GetSourceIndex) { + + const auto *VT = Call->getType()->getAs<VectorType>(); + if (!VT) + return false; + + APSInt MaskImm; + if (!EvaluateInteger(Call->getArg(2), MaskImm, Info)) + return false; + unsigned ShuffleMask = static_cast<unsigned>(MaskImm.getZExtValue()); + + APValue A, B; + if (!EvaluateAsRValue(Info, Call->getArg(0), A) || + !EvaluateAsRValue(Info, Call->getArg(1), B)) + return false; + + unsigned NumElts = VT->getNumElements(); + SmallVector<APValue, 16> ResultElements; + ResultElements.reserve(NumElts); + + for (unsigned DstIdx = 0; DstIdx != NumElts; ++DstIdx) { + auto [SrcVecIdx, SrcIdx] = GetSourceIndex(DstIdx, ShuffleMask); + const APValue &Src = (SrcVecIdx == 0) ? A : B; + ResultElements.push_back(Src.getVectorElt(SrcIdx)); + } + + Out = APValue(ResultElements.data(), ResultElements.size()); + return true; +} + static bool evalPshufbBuiltin(EvalInfo &Info, const CallExpr *Call, APValue &Out) { APValue SrcVec, ControlVec; @@ -12398,7 +12431,56 @@ bool VectorExprEvaluator::VisitCallExpr(const CallExpr *E) { return Success(APValue(ResultElements.data(), ResultElements.size()), E); } - + case X86::BI__builtin_ia32_shufps: + case X86::BI__builtin_ia32_shufps256: + case X86::BI__builtin_ia32_shufps512: { + APValue R; + if (!evalShuffleGeneric( + Info, E, R, + [](unsigned DstIdx, + unsigned ShuffleMask) -> std::pair<unsigned, unsigned> { + constexpr unsigned LaneBits = 128u; + unsigned NumElemPerLane = LaneBits / 32; + unsigned NumSelectableElems = NumElemPerLane / 2; + unsigned BitsPerElem = 2; + unsigned IndexMask = (1u << BitsPerElem) - 1; + unsigned MaskBits = 8; + unsigned Lane = DstIdx / NumElemPerLane; + unsigned ElemInLane = DstIdx % NumElemPerLane; + unsigned LaneOffset = Lane * NumElemPerLane; + unsigned BitIndex = (DstIdx * BitsPerElem) % MaskBits; + unsigned SrcIdx = (ElemInLane < NumSelectableElems) ? 0 : 1; + unsigned Index = (ShuffleMask >> BitIndex) & IndexMask; + return {SrcIdx, LaneOffset + Index}; + })) + return false; + return Success(R, E); + } + case X86::BI__builtin_ia32_shufpd: + case X86::BI__builtin_ia32_shufpd256: + case X86::BI__builtin_ia32_shufpd512: { + APValue R; + if (!evalShuffleGeneric( + Info, E, R, + [](unsigned DstIdx, + unsigned ShuffleMask) -> std::pair<unsigned, unsigned> { + constexpr unsigned LaneBits = 128u; + unsigned NumElemPerLane = LaneBits / 64; + unsigned NumSelectableElems = NumElemPerLane / 2; + unsigned BitsPerElem = 1; + unsigned IndexMask = (1u << BitsPerElem) - 1; + unsigned MaskBits = 8; + unsigned Lane = DstIdx / NumElemPerLane; + unsigned ElemInLane = DstIdx % NumElemPerLane; + unsigned LaneOffset = Lane * NumElemPerLane; + unsigned BitIndex = (DstIdx * BitsPerElem) % MaskBits; + unsigned SrcIdx = (ElemInLane < NumSelectableElems) ? 0 : 1; + unsigned Index = (ShuffleMask >> BitIndex) & IndexMask; + return {SrcIdx, LaneOffset + Index}; + })) + return false; + return Success(R, E); + } case X86::BI__builtin_ia32_pshufb128: case X86::BI__builtin_ia32_pshufb256: case X86::BI__builtin_ia32_pshufb512: { @@ -12906,6 +12988,66 @@ bool VectorExprEvaluator::VisitCallExpr(const CallExpr *E) { return Success(APValue(Elems.data(), NumElems), E); } + + case X86::BI__builtin_ia32_pslldqi128_byteshift: + case X86::BI__builtin_ia32_pslldqi256_byteshift: + case X86::BI__builtin_ia32_pslldqi512_byteshift: { + assert(E->getNumArgs() == 2); + + APValue Src; + APSInt Imm; + if (!EvaluateAsRValue(Info, E->getArg(0), Src) || + !EvaluateInteger(E->getArg(1), Imm, Info)) + return false; + + unsigned VecLen = Src.getVectorLength(); + unsigned Shift = Imm.getZExtValue() & 0xff; + + SmallVector<APValue> ResultElements; + for (unsigned Lane = 0; Lane != VecLen; Lane += 16) { + for (unsigned I = 0; I != 16; ++I) { + if (I < Shift) { + APSInt Zero(8, /*isUnsigned=*/true); + Zero = 0; + ResultElements.push_back(APValue(Zero)); + } else { + ResultElements.push_back(Src.getVectorElt(Lane + I - Shift)); + } + } + } + + return Success(APValue(ResultElements.data(), ResultElements.size()), E); + } + + case X86::BI__builtin_ia32_psrldqi128_byteshift: + case X86::BI__builtin_ia32_psrldqi256_byteshift: + case X86::BI__builtin_ia32_psrldqi512_byteshift: { + assert(E->getNumArgs() == 2); + + APValue Src; + APSInt Imm; + if (!EvaluateAsRValue(Info, E->getArg(0), Src) || + !EvaluateInteger(E->getArg(1), Imm, Info)) + return false; + + unsigned VecLen = Src.getVectorLength(); + unsigned Shift = Imm.getZExtValue() & 0xff; + + SmallVector<APValue> ResultElements; + for (unsigned Lane = 0; Lane != VecLen; Lane += 16) { + for (unsigned I = 0; I != 16; ++I) { + if (I + Shift < 16) { + ResultElements.push_back(Src.getVectorElt(Lane + I + Shift)); + } else { + APSInt Zero(8, /*isUnsigned=*/true); + Zero = 0; + ResultElements.push_back(APValue(Zero)); + } + } + } + + return Success(APValue(ResultElements.data(), ResultElements.size()), E); + } } } diff --git a/clang/lib/AST/StmtOpenACC.cpp b/clang/lib/AST/StmtOpenACC.cpp index 462a10d..39dfa19 100644 --- a/clang/lib/AST/StmtOpenACC.cpp +++ b/clang/lib/AST/StmtOpenACC.cpp @@ -326,16 +326,30 @@ OpenACCAtomicConstruct *OpenACCAtomicConstruct::Create( static std::pair<const Expr *, const Expr *> getBinaryOpArgs(const Expr *Op) { if (const auto *BO = dyn_cast<BinaryOperator>(Op)) { - assert(BO->getOpcode() == BO_Assign); + assert(BO->isAssignmentOp()); return {BO->getLHS(), BO->getRHS()}; } const auto *OO = cast<CXXOperatorCallExpr>(Op); - assert(OO->getOperator() == OO_Equal); - + assert(OO->isAssignmentOp()); return {OO->getArg(0), OO->getArg(1)}; } +static std::pair<bool, const Expr *> getUnaryOpArgs(const Expr *Op) { + if (const auto *UO = dyn_cast<UnaryOperator>(Op)) + return {true, UO->getSubExpr()}; + + if (const auto *OpCall = dyn_cast<CXXOperatorCallExpr>(Op)) { + // Post-inc/dec have a second unused argument to differentiate it, so we + // accept -- or ++ as unary, or any operator call with only 1 arg. + if (OpCall->getNumArgs() == 1 || OpCall->getOperator() != OO_PlusPlus || + OpCall->getOperator() != OO_MinusMinus) + return {true, OpCall->getArg(0)}; + } + + return {false, nullptr}; +} + const OpenACCAtomicConstruct::StmtInfo OpenACCAtomicConstruct::getAssociatedStmtInfo() const { // This ends up being a vastly simplified version of SemaOpenACCAtomic, since @@ -343,18 +357,17 @@ OpenACCAtomicConstruct::getAssociatedStmtInfo() const { // asserts to ensure we don't get off into the weeds. assert(getAssociatedStmt() && "invalid associated stmt?"); + const Expr *AssocStmt = cast<const Expr>(getAssociatedStmt()); switch (AtomicKind) { - case OpenACCAtomicKind::None: - case OpenACCAtomicKind::Update: case OpenACCAtomicKind::Capture: - assert(false && "Only 'read'/'write' have been implemented here"); + assert(false && "Only 'read'/'write'/'update' have been implemented here"); return {}; case OpenACCAtomicKind::Read: { // Read only supports the format 'v = x'; where both sides are a scalar // expression. This can come in 2 forms; BinaryOperator or // CXXOperatorCallExpr (rarely). std::pair<const Expr *, const Expr *> BinaryArgs = - getBinaryOpArgs(cast<const Expr>(getAssociatedStmt())); + getBinaryOpArgs(AssocStmt); // We want the L-value for each side, so we ignore implicit casts. return {BinaryArgs.first->IgnoreImpCasts(), BinaryArgs.second->IgnoreImpCasts(), /*expr=*/nullptr}; @@ -364,13 +377,28 @@ OpenACCAtomicConstruct::getAssociatedStmtInfo() const { // type, and 'x' is a scalar l value. As above, this can come in 2 forms; // Binary Operator or CXXOperatorCallExpr. std::pair<const Expr *, const Expr *> BinaryArgs = - getBinaryOpArgs(cast<const Expr>(getAssociatedStmt())); + getBinaryOpArgs(AssocStmt); // We want the L-value for ONLY the X side, so we ignore implicit casts. For // the right side (the expr), we emit it as an r-value so we need to // maintain implicit casts. return {/*v=*/nullptr, BinaryArgs.first->IgnoreImpCasts(), BinaryArgs.second}; } + case OpenACCAtomicKind::None: + case OpenACCAtomicKind::Update: { + std::pair<bool, const Expr *> UnaryArgs = getUnaryOpArgs(AssocStmt); + if (UnaryArgs.first) + return {/*v=*/nullptr, UnaryArgs.second->IgnoreImpCasts(), + /*expr=*/nullptr}; + + std::pair<const Expr *, const Expr *> BinaryArgs = + getBinaryOpArgs(AssocStmt); + // For binary args, we just store the RHS as an expression (in the + // expression slot), since the codegen just wants the whole thing for a + // recipe. + return {/*v=*/nullptr, BinaryArgs.first->IgnoreImpCasts(), + BinaryArgs.second}; + } } llvm_unreachable("unknown OpenACC atomic kind"); diff --git a/clang/lib/Basic/FileManager.cpp b/clang/lib/Basic/FileManager.cpp index 7481e1e..e744cc0 100644 --- a/clang/lib/Basic/FileManager.cpp +++ b/clang/lib/Basic/FileManager.cpp @@ -474,8 +474,9 @@ OptionalFileEntryRef FileManager::getBypassFile(FileEntryRef VF) { return FileEntryRef(*Insertion.first); } -bool FileManager::FixupRelativePath(SmallVectorImpl<char> &path) const { - StringRef pathRef(path.data(), path.size()); +bool FileManager::fixupRelativePath(const FileSystemOptions &FileSystemOpts, + SmallVectorImpl<char> &Path) { + StringRef pathRef(Path.data(), Path.size()); if (FileSystemOpts.WorkingDir.empty() || llvm::sys::path::is_absolute(pathRef)) @@ -483,7 +484,7 @@ bool FileManager::FixupRelativePath(SmallVectorImpl<char> &path) const { SmallString<128> NewPath(FileSystemOpts.WorkingDir); llvm::sys::path::append(NewPath, pathRef); - path = NewPath; + Path = NewPath; return true; } diff --git a/clang/lib/Basic/Targets/AArch64.cpp b/clang/lib/Basic/Targets/AArch64.cpp index 146f058..c2d1bc1 100644 --- a/clang/lib/Basic/Targets/AArch64.cpp +++ b/clang/lib/Basic/Targets/AArch64.cpp @@ -206,8 +206,7 @@ AArch64TargetInfo::AArch64TargetInfo(const llvm::Triple &Triple, StringRef AArch64TargetInfo::getABI() const { return ABI; } bool AArch64TargetInfo::setABI(const std::string &Name) { - if (Name != "aapcs" && Name != "aapcs-soft" && Name != "darwinpcs" && - Name != "pauthtest") + if (Name != "aapcs" && Name != "aapcs-soft" && Name != "darwinpcs") return false; ABI = Name; @@ -221,12 +220,6 @@ bool AArch64TargetInfo::validateTarget(DiagnosticsEngine &Diags) const { Diags.Report(diag::err_target_unsupported_abi_with_fpu) << ABI; return false; } - if (getTriple().getEnvironment() == llvm::Triple::PAuthTest && - getTriple().getOS() != llvm::Triple::Linux) { - Diags.Report(diag::err_target_unsupported_abi_for_triple) - << getTriple().getEnvironmentName() << getTriple().getTriple(); - return false; - } return true; } diff --git a/clang/lib/Basic/Targets/AArch64.h b/clang/lib/Basic/Targets/AArch64.h index 3952e7b..7d0737b 100644 --- a/clang/lib/Basic/Targets/AArch64.h +++ b/clang/lib/Basic/Targets/AArch64.h @@ -135,6 +135,7 @@ class LLVM_LIBRARY_VISIBILITY AArch64TargetInfo : public TargetInfo { const llvm::AArch64::ArchInfo *ArchInfo = &llvm::AArch64::ARMV8A; +protected: std::string ABI; public: @@ -279,6 +280,16 @@ private: void setDataLayout() override; }; +template <> +inline bool +LinuxTargetInfo<AArch64leTargetInfo>::setABI(const std::string &Name) { + if (Name == "pauthtest") { + ABI = Name; + return true; + } + return AArch64leTargetInfo::setABI(Name); +} + class LLVM_LIBRARY_VISIBILITY WindowsARM64TargetInfo : public WindowsTargetInfo<AArch64leTargetInfo> { const llvm::Triple Triple; diff --git a/clang/lib/Basic/Targets/Hexagon.cpp b/clang/lib/Basic/Targets/Hexagon.cpp index cea64f9..d5b413cb 100644 --- a/clang/lib/Basic/Targets/Hexagon.cpp +++ b/clang/lib/Basic/Targets/Hexagon.cpp @@ -83,6 +83,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions &Opts, } else if (CPU == "hexagonv79") { Builder.defineMacro("__HEXAGON_V79__"); Builder.defineMacro("__HEXAGON_ARCH__", "79"); + } else if (CPU == "hexagonv81") { + Builder.defineMacro("__HEXAGON_V81__"); + Builder.defineMacro("__HEXAGON_ARCH__", "81"); } if (hasFeature("hvx-length64b")) { @@ -252,8 +255,7 @@ static constexpr CPUSuffix Suffixes[] = { {{"hexagonv68"}, {"68"}}, {{"hexagonv69"}, {"69"}}, {{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}}, {{"hexagonv73"}, {"73"}}, {{"hexagonv75"}, {"75"}}, - {{"hexagonv79"}, {"79"}}, -}; + {{"hexagonv79"}, {"79"}}, {{"hexagonv81"}, {"81"}}}; std::optional<unsigned> HexagonTargetInfo::getHexagonCPURev(StringRef Name) { StringRef Arch = Name; diff --git a/clang/lib/Basic/Targets/OSTargets.cpp b/clang/lib/Basic/Targets/OSTargets.cpp index e744e84..e99bbd1 100644 --- a/clang/lib/Basic/Targets/OSTargets.cpp +++ b/clang/lib/Basic/Targets/OSTargets.cpp @@ -10,6 +10,7 @@ //===----------------------------------------------------------------------===// #include "OSTargets.h" +#include "AArch64.h" #include "clang/Basic/MacroBuilder.h" #include "llvm/ADT/StringRef.h" diff --git a/clang/lib/Basic/Targets/OSTargets.h b/clang/lib/Basic/Targets/OSTargets.h index bd6ffcf..4d81c9a 100644 --- a/clang/lib/Basic/Targets/OSTargets.h +++ b/clang/lib/Basic/Targets/OSTargets.h @@ -408,6 +408,12 @@ public: const char *getStaticInitSectionSpecifier() const override { return ".text.startup"; } + + // This allows template specializations, see + // LinuxTargetInfo<AArch64leTargetInfo>::setABI + bool setABI(const std::string &Name) override { + return OSTargetInfo<Target>::setABI(Name); + } }; // Managarm Target diff --git a/clang/lib/CIR/CodeGen/CIRGenBuiltin.cpp b/clang/lib/CIR/CodeGen/CIRGenBuiltin.cpp index 62fa04e..e35100f 100644 --- a/clang/lib/CIR/CodeGen/CIRGenBuiltin.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenBuiltin.cpp @@ -449,10 +449,15 @@ RValue CIRGenFunction::emitBuiltinExpr(const GlobalDecl &gd, unsigned builtinID, } case Builtin::BI__builtin_coro_free: case Builtin::BI__builtin_coro_size: { - cgm.errorNYI(e->getSourceRange(), - "BI__builtin_coro_free, BI__builtin_coro_size NYI"); - assert(!cir::MissingFeatures::coroSizeBuiltinCall()); - return getUndefRValue(e->getType()); + GlobalDecl gd{fd}; + mlir::Type ty = cgm.getTypes().getFunctionType( + cgm.getTypes().arrangeGlobalDeclaration(gd)); + const auto *nd = cast<NamedDecl>(gd.getDecl()); + cir::FuncOp fnOp = + cgm.getOrCreateCIRFunction(nd->getName(), ty, gd, /*ForVTable=*/false); + fnOp.setBuiltin(true); + return emitCall(e->getCallee()->getType(), CIRGenCallee::forDirect(fnOp), e, + returnValue); } case Builtin::BI__builtin_prefetch: { auto evaluateOperandAsInt = [&](const Expr *arg) { diff --git a/clang/lib/CIR/CodeGen/CIRGenCXXABI.h b/clang/lib/CIR/CodeGen/CIRGenCXXABI.h index d3c7dac0..13dc9f3 100644 --- a/clang/lib/CIR/CodeGen/CIRGenCXXABI.h +++ b/clang/lib/CIR/CodeGen/CIRGenCXXABI.h @@ -187,6 +187,11 @@ public: virtual void registerGlobalDtor(const VarDecl *vd, cir::FuncOp dtor, mlir::Value addr) = 0; + virtual void emitVirtualObjectDelete(CIRGenFunction &cgf, + const CXXDeleteExpr *de, Address ptr, + QualType elementType, + const CXXDestructorDecl *dtor) = 0; + /// Checks if ABI requires extra virtual offset for vtable field. virtual bool isVirtualOffsetNeededForVTableField(CIRGenFunction &cgf, diff --git a/clang/lib/CIR/CodeGen/CIRGenCall.cpp b/clang/lib/CIR/CodeGen/CIRGenCall.cpp index 88aef89..50d4c03 100644 --- a/clang/lib/CIR/CodeGen/CIRGenCall.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenCall.cpp @@ -671,9 +671,12 @@ RValue CIRGenFunction::emitCall(const CIRGenFunctionInfo &funcInfo, return RValue::get(results[0]); } - case cir::TEK_Complex: - cgm.errorNYI(loc, "unsupported evaluation kind of function call result"); - return getUndefRValue(retTy); + case cir::TEK_Complex: { + mlir::ResultRange results = theCall->getOpResults(); + assert(!results.empty() && + "Expected at least one result for complex rvalue"); + return RValue::getComplex(results[0]); + } } llvm_unreachable("Invalid evaluation kind"); } diff --git a/clang/lib/CIR/CodeGen/CIRGenCleanup.h b/clang/lib/CIR/CodeGen/CIRGenCleanup.h index 9acf8b1..61a09a5 100644 --- a/clang/lib/CIR/CodeGen/CIRGenCleanup.h +++ b/clang/lib/CIR/CodeGen/CIRGenCleanup.h @@ -15,6 +15,7 @@ #define CLANG_LIB_CIR_CODEGEN_CIRGENCLEANUP_H #include "Address.h" +#include "CIRGenModule.h" #include "EHScopeStack.h" #include "mlir/IR/Value.h" @@ -257,5 +258,53 @@ inline void EHScopeStack::popCatch() { deallocate(EHCatchScope::getSizeForNumHandlers(scope.getNumHandlers())); } +/// The exceptions personality for a function. +struct EHPersonality { + const char *personalityFn = nullptr; + + // If this is non-null, this personality requires a non-standard + // function for rethrowing an exception after a catchall cleanup. + // This function must have prototype void(void*). + const char *catchallRethrowFn = nullptr; + + static const EHPersonality &get(CIRGenModule &cgm, + const clang::FunctionDecl *fd); + static const EHPersonality &get(CIRGenFunction &cgf); + + static const EHPersonality GNU_C; + static const EHPersonality GNU_C_SJLJ; + static const EHPersonality GNU_C_SEH; + static const EHPersonality GNU_ObjC; + static const EHPersonality GNU_ObjC_SJLJ; + static const EHPersonality GNU_ObjC_SEH; + static const EHPersonality GNUstep_ObjC; + static const EHPersonality GNU_ObjCXX; + static const EHPersonality NeXT_ObjC; + static const EHPersonality GNU_CPlusPlus; + static const EHPersonality GNU_CPlusPlus_SJLJ; + static const EHPersonality GNU_CPlusPlus_SEH; + static const EHPersonality MSVC_except_handler; + static const EHPersonality MSVC_C_specific_handler; + static const EHPersonality MSVC_CxxFrameHandler3; + static const EHPersonality GNU_Wasm_CPlusPlus; + static const EHPersonality XL_CPlusPlus; + static const EHPersonality ZOS_CPlusPlus; + + /// Does this personality use landingpads or the family of pad instructions + /// designed to form funclets? + bool usesFuncletPads() const { + return isMSVCPersonality() || isWasmPersonality(); + } + + bool isMSVCPersonality() const { + return this == &MSVC_except_handler || this == &MSVC_C_specific_handler || + this == &MSVC_CxxFrameHandler3; + } + + bool isWasmPersonality() const { return this == &GNU_Wasm_CPlusPlus; } + + bool isMSVCXXPersonality() const { return this == &MSVC_CxxFrameHandler3; } +}; + } // namespace clang::CIRGen #endif // CLANG_LIB_CIR_CODEGEN_CIRGENCLEANUP_H diff --git a/clang/lib/CIR/CodeGen/CIRGenCoroutine.cpp b/clang/lib/CIR/CodeGen/CIRGenCoroutine.cpp index c25cce4..8723a6e 100644 --- a/clang/lib/CIR/CodeGen/CIRGenCoroutine.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenCoroutine.cpp @@ -15,6 +15,7 @@ #include "clang/AST/StmtCXX.h" #include "clang/Basic/TargetInfo.h" #include "clang/CIR/Dialect/IR/CIRTypes.h" +#include "clang/CIR/MissingFeatures.h" using namespace clang; using namespace clang::CIRGen; @@ -23,6 +24,9 @@ struct clang::CIRGen::CGCoroData { // Stores the __builtin_coro_id emitted in the function so that we can supply // it as the first argument to other builtins. cir::CallOp coroId = nullptr; + + // Stores the result of __builtin_coro_begin call. + mlir::Value coroBegin = nullptr; }; // Defining these here allows to keep CGCoroData private to this file. @@ -63,6 +67,46 @@ cir::CallOp CIRGenFunction::emitCoroIDBuiltinCall(mlir::Location loc, nullPtr, nullPtr, nullPtr}); } +cir::CallOp CIRGenFunction::emitCoroAllocBuiltinCall(mlir::Location loc) { + cir::BoolType boolTy = builder.getBoolTy(); + + mlir::Operation *builtin = cgm.getGlobalValue(cgm.builtinCoroAlloc); + + cir::FuncOp fnOp; + if (!builtin) { + fnOp = cgm.createCIRBuiltinFunction(loc, cgm.builtinCoroAlloc, + cir::FuncType::get({UInt32Ty}, boolTy), + /*fd=*/nullptr); + assert(fnOp && "should always succeed"); + } else { + fnOp = cast<cir::FuncOp>(builtin); + } + + return builder.createCallOp( + loc, fnOp, mlir::ValueRange{curCoro.data->coroId.getResult()}); +} + +cir::CallOp +CIRGenFunction::emitCoroBeginBuiltinCall(mlir::Location loc, + mlir::Value coroframeAddr) { + mlir::Operation *builtin = cgm.getGlobalValue(cgm.builtinCoroBegin); + + cir::FuncOp fnOp; + if (!builtin) { + fnOp = cgm.createCIRBuiltinFunction( + loc, cgm.builtinCoroBegin, + cir::FuncType::get({UInt32Ty, VoidPtrTy}, VoidPtrTy), + /*fd=*/nullptr); + assert(fnOp && "should always succeed"); + } else { + fnOp = cast<cir::FuncOp>(builtin); + } + + return builder.createCallOp( + loc, fnOp, + mlir::ValueRange{curCoro.data->coroId.getResult(), coroframeAddr}); +} + mlir::LogicalResult CIRGenFunction::emitCoroutineBody(const CoroutineBodyStmt &s) { mlir::Location openCurlyLoc = getLoc(s.getBeginLoc()); @@ -73,10 +117,39 @@ CIRGenFunction::emitCoroutineBody(const CoroutineBodyStmt &s) { cir::CallOp coroId = emitCoroIDBuiltinCall(openCurlyLoc, nullPtrCst); createCoroData(*this, curCoro, coroId); - assert(!cir::MissingFeatures::coroAllocBuiltinCall()); - - assert(!cir::MissingFeatures::coroBeginBuiltinCall()); + // Backend is allowed to elide memory allocations, to help it, emit + // auto mem = coro.alloc() ? 0 : ... allocation code ...; + cir::CallOp coroAlloc = emitCoroAllocBuiltinCall(openCurlyLoc); + + // Initialize address of coroutine frame to null + CanQualType astVoidPtrTy = cgm.getASTContext().VoidPtrTy; + mlir::Type allocaTy = convertTypeForMem(astVoidPtrTy); + Address coroFrame = + createTempAlloca(allocaTy, getContext().getTypeAlignInChars(astVoidPtrTy), + openCurlyLoc, "__coro_frame_addr", + /*ArraySize=*/nullptr); + + mlir::Value storeAddr = coroFrame.getPointer(); + builder.CIRBaseBuilderTy::createStore(openCurlyLoc, nullPtrCst, storeAddr); + cir::IfOp::create( + builder, openCurlyLoc, coroAlloc.getResult(), + /*withElseRegion=*/false, + /*thenBuilder=*/[&](mlir::OpBuilder &b, mlir::Location loc) { + builder.CIRBaseBuilderTy::createStore( + loc, emitScalarExpr(s.getAllocate()), storeAddr); + cir::YieldOp::create(builder, loc); + }); + curCoro.data->coroBegin = + emitCoroBeginBuiltinCall( + openCurlyLoc, + cir::LoadOp::create(builder, openCurlyLoc, allocaTy, storeAddr)) + .getResult(); + + // Handle allocation failure if 'ReturnStmtOnAllocFailure' was provided. + if (s.getReturnStmtOnAllocFailure()) + cgm.errorNYI("handle coroutine return alloc failure"); assert(!cir::MissingFeatures::generateDebugInfo()); + assert(!cir::MissingFeatures::emitBodyAndFallthrough()); return mlir::success(); } diff --git a/clang/lib/CIR/CodeGen/CIRGenException.cpp b/clang/lib/CIR/CodeGen/CIRGenException.cpp index 717a3e0..67f46ff 100644 --- a/clang/lib/CIR/CodeGen/CIRGenException.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenException.cpp @@ -18,6 +18,171 @@ using namespace clang; using namespace clang::CIRGen; +const EHPersonality EHPersonality::GNU_C = {"__gcc_personality_v0", nullptr}; +const EHPersonality EHPersonality::GNU_C_SJLJ = {"__gcc_personality_sj0", + nullptr}; +const EHPersonality EHPersonality::GNU_C_SEH = {"__gcc_personality_seh0", + nullptr}; +const EHPersonality EHPersonality::NeXT_ObjC = {"__objc_personality_v0", + nullptr}; +const EHPersonality EHPersonality::GNU_CPlusPlus = {"__gxx_personality_v0", + nullptr}; +const EHPersonality EHPersonality::GNU_CPlusPlus_SJLJ = { + "__gxx_personality_sj0", nullptr}; +const EHPersonality EHPersonality::GNU_CPlusPlus_SEH = { + "__gxx_personality_seh0", nullptr}; +const EHPersonality EHPersonality::GNU_ObjC = {"__gnu_objc_personality_v0", + "objc_exception_throw"}; +const EHPersonality EHPersonality::GNU_ObjC_SJLJ = { + "__gnu_objc_personality_sj0", "objc_exception_throw"}; +const EHPersonality EHPersonality::GNU_ObjC_SEH = { + "__gnu_objc_personality_seh0", "objc_exception_throw"}; +const EHPersonality EHPersonality::GNU_ObjCXX = { + "__gnustep_objcxx_personality_v0", nullptr}; +const EHPersonality EHPersonality::GNUstep_ObjC = { + "__gnustep_objc_personality_v0", nullptr}; +const EHPersonality EHPersonality::MSVC_except_handler = {"_except_handler3", + nullptr}; +const EHPersonality EHPersonality::MSVC_C_specific_handler = { + "__C_specific_handler", nullptr}; +const EHPersonality EHPersonality::MSVC_CxxFrameHandler3 = { + "__CxxFrameHandler3", nullptr}; +const EHPersonality EHPersonality::GNU_Wasm_CPlusPlus = { + "__gxx_wasm_personality_v0", nullptr}; +const EHPersonality EHPersonality::XL_CPlusPlus = {"__xlcxx_personality_v1", + nullptr}; +const EHPersonality EHPersonality::ZOS_CPlusPlus = {"__zos_cxx_personality_v2", + nullptr}; + +static const EHPersonality &getCPersonality(const TargetInfo &target, + const CodeGenOptions &cgOpts) { + const llvm::Triple &triple = target.getTriple(); + if (triple.isWindowsMSVCEnvironment()) + return EHPersonality::MSVC_CxxFrameHandler3; + if (cgOpts.hasSjLjExceptions()) + return EHPersonality::GNU_C_SJLJ; + if (cgOpts.hasDWARFExceptions()) + return EHPersonality::GNU_C; + if (cgOpts.hasSEHExceptions()) + return EHPersonality::GNU_C_SEH; + return EHPersonality::GNU_C; +} + +static const EHPersonality &getObjCPersonality(const TargetInfo &target, + const LangOptions &langOpts, + const CodeGenOptions &cgOpts) { + const llvm::Triple &triple = target.getTriple(); + if (triple.isWindowsMSVCEnvironment()) + return EHPersonality::MSVC_CxxFrameHandler3; + + switch (langOpts.ObjCRuntime.getKind()) { + case ObjCRuntime::FragileMacOSX: + return getCPersonality(target, cgOpts); + case ObjCRuntime::MacOSX: + case ObjCRuntime::iOS: + case ObjCRuntime::WatchOS: + return EHPersonality::NeXT_ObjC; + case ObjCRuntime::GNUstep: + if (langOpts.ObjCRuntime.getVersion() >= VersionTuple(1, 7)) + return EHPersonality::GNUstep_ObjC; + [[fallthrough]]; + case ObjCRuntime::GCC: + case ObjCRuntime::ObjFW: + if (cgOpts.hasSjLjExceptions()) + return EHPersonality::GNU_ObjC_SJLJ; + if (cgOpts.hasSEHExceptions()) + return EHPersonality::GNU_ObjC_SEH; + return EHPersonality::GNU_ObjC; + } + llvm_unreachable("bad runtime kind"); +} + +static const EHPersonality &getCXXPersonality(const TargetInfo &target, + const CodeGenOptions &cgOpts) { + const llvm::Triple &triple = target.getTriple(); + if (triple.isWindowsMSVCEnvironment()) + return EHPersonality::MSVC_CxxFrameHandler3; + if (triple.isOSAIX()) + return EHPersonality::XL_CPlusPlus; + if (cgOpts.hasSjLjExceptions()) + return EHPersonality::GNU_CPlusPlus_SJLJ; + if (cgOpts.hasDWARFExceptions()) + return EHPersonality::GNU_CPlusPlus; + if (cgOpts.hasSEHExceptions()) + return EHPersonality::GNU_CPlusPlus_SEH; + if (cgOpts.hasWasmExceptions()) + return EHPersonality::GNU_Wasm_CPlusPlus; + return EHPersonality::GNU_CPlusPlus; +} + +/// Determines the personality function to use when both C++ +/// and Objective-C exceptions are being caught. +static const EHPersonality &getObjCXXPersonality(const TargetInfo &target, + const LangOptions &langOpts, + const CodeGenOptions &cgOpts) { + if (target.getTriple().isWindowsMSVCEnvironment()) + return EHPersonality::MSVC_CxxFrameHandler3; + + switch (langOpts.ObjCRuntime.getKind()) { + // In the fragile ABI, just use C++ exception handling and hope + // they're not doing crazy exception mixing. + case ObjCRuntime::FragileMacOSX: + return getCXXPersonality(target, cgOpts); + + // The ObjC personality defers to the C++ personality for non-ObjC + // handlers. Unlike the C++ case, we use the same personality + // function on targets using (backend-driven) SJLJ EH. + case ObjCRuntime::MacOSX: + case ObjCRuntime::iOS: + case ObjCRuntime::WatchOS: + return getObjCPersonality(target, langOpts, cgOpts); + + case ObjCRuntime::GNUstep: + return EHPersonality::GNU_ObjCXX; + + // The GCC runtime's personality function inherently doesn't support + // mixed EH. Use the ObjC personality just to avoid returning null. + case ObjCRuntime::GCC: + case ObjCRuntime::ObjFW: + return getObjCPersonality(target, langOpts, cgOpts); + } + llvm_unreachable("bad runtime kind"); +} + +static const EHPersonality &getSEHPersonalityMSVC(const llvm::Triple &triple) { + return triple.getArch() == llvm::Triple::x86 + ? EHPersonality::MSVC_except_handler + : EHPersonality::MSVC_C_specific_handler; +} + +const EHPersonality &EHPersonality::get(CIRGenModule &cgm, + const FunctionDecl *fd) { + const llvm::Triple &triple = cgm.getTarget().getTriple(); + const LangOptions &langOpts = cgm.getLangOpts(); + const CodeGenOptions &cgOpts = cgm.getCodeGenOpts(); + const TargetInfo &target = cgm.getTarget(); + + // Functions using SEH get an SEH personality. + if (fd && fd->usesSEHTry()) + return getSEHPersonalityMSVC(triple); + + if (langOpts.ObjC) { + return langOpts.CPlusPlus ? getObjCXXPersonality(target, langOpts, cgOpts) + : getObjCPersonality(target, langOpts, cgOpts); + } + return langOpts.CPlusPlus ? getCXXPersonality(target, cgOpts) + : getCPersonality(target, cgOpts); +} + +const EHPersonality &EHPersonality::get(CIRGenFunction &cgf) { + const auto *fg = cgf.curCodeDecl; + // For outlined finallys and filters, use the SEH personality in case they + // contain more SEH. This mostly only affects finallys. Filters could + // hypothetically use gnu statement expressions to sneak in nested SEH. + fg = fg ? fg : cgf.curSEHParent.getDecl(); + return get(cgf.cgm, dyn_cast_or_null<FunctionDecl>(fg)); +} + void CIRGenFunction::emitCXXThrowExpr(const CXXThrowExpr *e) { const llvm::Triple &triple = getTarget().getTriple(); if (cgm.getLangOpts().OpenMPIsTargetDevice && diff --git a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp index d6d226b..8fe0d9b4 100644 --- a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp @@ -362,8 +362,7 @@ public: cgf.cgm.errorNYI(e->getSourceRange(), "AggExprEmitter: VisitCXXTypeidExpr"); } void VisitMaterializeTemporaryExpr(MaterializeTemporaryExpr *e) { - cgf.cgm.errorNYI(e->getSourceRange(), - "AggExprEmitter: VisitMaterializeTemporaryExpr"); + Visit(e->getSubExpr()); } void VisitOpaqueValueExpr(OpaqueValueExpr *e) { cgf.cgm.errorNYI(e->getSourceRange(), diff --git a/clang/lib/CIR/CodeGen/CIRGenExprCXX.cpp b/clang/lib/CIR/CodeGen/CIRGenExprCXX.cpp index a3cdf19..7a35382 100644 --- a/clang/lib/CIR/CodeGen/CIRGenExprCXX.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenExprCXX.cpp @@ -565,8 +565,10 @@ static void emitObjectDelete(CIRGenFunction &cgf, const CXXDeleteExpr *de, dtor = rd->getDestructor(); if (dtor->isVirtual()) { - cgf.cgm.errorNYI(de->getSourceRange(), - "emitObjectDelete: virtual destructor"); + assert(!cir::MissingFeatures::devirtualizeDestructor()); + cgf.cgm.getCXXABI().emitVirtualObjectDelete(cgf, de, ptr, elementType, + dtor); + return; } } } diff --git a/clang/lib/CIR/CodeGen/CIRGenExprConstant.cpp b/clang/lib/CIR/CodeGen/CIRGenExprConstant.cpp index 7de3dd0..928e5aa 100644 --- a/clang/lib/CIR/CodeGen/CIRGenExprConstant.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenExprConstant.cpp @@ -922,9 +922,9 @@ public: } mlir::Attribute VisitCastExpr(CastExpr *e, QualType destType) { - if (isa<ExplicitCastExpr>(e)) - cgm.errorNYI(e->getBeginLoc(), - "ConstExprEmitter::VisitCastExpr explicit cast"); + if (const auto *ece = dyn_cast<ExplicitCastExpr>(e)) + cgm.emitExplicitCastExprType(ece, + const_cast<CIRGenFunction *>(emitter.cgf)); Expr *subExpr = e->getSubExpr(); @@ -1078,9 +1078,32 @@ public: mlir::Attribute VisitCXXConstructExpr(CXXConstructExpr *e, QualType ty) { if (!e->getConstructor()->isTrivial()) - return nullptr; - cgm.errorNYI(e->getBeginLoc(), "trivial constructor const handling"); - return {}; + return {}; + + // Only default and copy/move constructors can be trivial. + if (e->getNumArgs()) { + assert(e->getNumArgs() == 1 && "trivial ctor with > 1 argument"); + assert(e->getConstructor()->isCopyOrMoveConstructor() && + "trivial ctor has argument but isn't a copy/move ctor"); + + Expr *arg = e->getArg(0); + assert(cgm.getASTContext().hasSameUnqualifiedType(ty, arg->getType()) && + "argument to copy ctor is of wrong type"); + + // Look through the temporary; it's just converting the value to an lvalue + // to pass it to the constructor. + if (auto const *mte = dyn_cast<MaterializeTemporaryExpr>(arg)) + return Visit(mte->getSubExpr(), ty); + + // TODO: Investigate whether there are cases that can fall through to here + // that need to be handled. This is missing in classic codegen also. + assert(!cir::MissingFeatures::ctorConstLvalueToRvalueConversion()); + + // Don't try to support arbitrary lvalue-to-rvalue conversions for now. + return {}; + } + + return cgm.getBuilder().getZeroInitAttr(cgm.convertType(ty)); } mlir::Attribute VisitStringLiteral(StringLiteral *e, QualType t) { diff --git a/clang/lib/CIR/CodeGen/CIRGenFunction.h b/clang/lib/CIR/CodeGen/CIRGenFunction.h index 5f9dbdc..c3fcd1a6 100644 --- a/clang/lib/CIR/CodeGen/CIRGenFunction.h +++ b/clang/lib/CIR/CodeGen/CIRGenFunction.h @@ -120,6 +120,8 @@ public: /// Tracks function scope overall cleanup handling. EHScopeStack ehStack; + GlobalDecl curSEHParent; + llvm::DenseMap<const clang::ValueDecl *, clang::FieldDecl *> lambdaCaptureFields; clang::FieldDecl *lambdaThisCaptureField = nullptr; @@ -665,6 +667,12 @@ public: symbolTable.insert(vd, addr.getPointer()); } + // Replaces the address of the local variable, if it exists. Else does the + // same thing as setAddrOfLocalVar. + void replaceAddrOfLocalVar(const clang::VarDecl *vd, Address addr) { + localDeclMap.insert_or_assign(vd, addr); + } + // A class to allow reverting changes to a var-decl's registration to the // localDeclMap. This is used in cases where things are being inserted into // the variable list but don't follow normal lookup/search rules, like in @@ -1326,6 +1334,9 @@ public: mlir::LogicalResult emitCoroutineBody(const CoroutineBodyStmt &s); cir::CallOp emitCoroEndBuiltinCall(mlir::Location loc, mlir::Value nullPtr); cir::CallOp emitCoroIDBuiltinCall(mlir::Location loc, mlir::Value nullPtr); + cir::CallOp emitCoroAllocBuiltinCall(mlir::Location loc); + cir::CallOp emitCoroBeginBuiltinCall(mlir::Location loc, + mlir::Value coroframeAddr); void emitDestroy(Address addr, QualType type, Destroyer *destroyer); diff --git a/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp b/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp index 2dce0b1..88fedf1 100644 --- a/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp @@ -74,6 +74,9 @@ public: QualType thisTy) override; void registerGlobalDtor(const VarDecl *vd, cir::FuncOp dtor, mlir::Value addr) override; + void emitVirtualObjectDelete(CIRGenFunction &cgf, const CXXDeleteExpr *de, + Address ptr, QualType elementType, + const CXXDestructorDecl *dtor) override; void emitRethrow(CIRGenFunction &cgf, bool isNoReturn) override; void emitThrow(CIRGenFunction &cgf, const CXXThrowExpr *e) override; @@ -2175,6 +2178,21 @@ mlir::Value CIRGenItaniumCXXABI::emitDynamicCast(CIRGenFunction &cgf, isRefCast, castInfo); } +/// The Itanium ABI always places an offset to the complete object +/// at entry -2 in the vtable. +void CIRGenItaniumCXXABI::emitVirtualObjectDelete( + CIRGenFunction &cgf, const CXXDeleteExpr *delExpr, Address ptr, + QualType elementType, const CXXDestructorDecl *dtor) { + bool useGlobalDelete = delExpr->isGlobalDelete(); + if (useGlobalDelete) { + cgf.cgm.errorNYI(delExpr->getSourceRange(), + "emitVirtualObjectDelete: global delete"); + } + + CXXDtorType dtorType = useGlobalDelete ? Dtor_Complete : Dtor_Deleting; + emitVirtualDestructorCall(cgf, dtor, dtorType, ptr, delExpr); +} + /************************** Array allocation cookies **************************/ CharUnits CIRGenItaniumCXXABI::getArrayCookieSizeImpl(QualType elementType) { diff --git a/clang/lib/CIR/CodeGen/CIRGenModule.h b/clang/lib/CIR/CodeGen/CIRGenModule.h index 1fc116d..186913d 100644 --- a/clang/lib/CIR/CodeGen/CIRGenModule.h +++ b/clang/lib/CIR/CodeGen/CIRGenModule.h @@ -496,6 +496,8 @@ public: bool assumeConvergent = false); static constexpr const char *builtinCoroId = "__builtin_coro_id"; + static constexpr const char *builtinCoroAlloc = "__builtin_coro_alloc"; + static constexpr const char *builtinCoroBegin = "__builtin_coro_begin"; /// Given a builtin id for a function like "__builtin_fabsf", return a /// Function* for "fabsf". diff --git a/clang/lib/CIR/CodeGen/CIRGenStmtOpenACC.cpp b/clang/lib/CIR/CodeGen/CIRGenStmtOpenACC.cpp index 349b111..9e55bd5 100644 --- a/clang/lib/CIR/CodeGen/CIRGenStmtOpenACC.cpp +++ b/clang/lib/CIR/CodeGen/CIRGenStmtOpenACC.cpp @@ -304,12 +304,21 @@ CIRGenFunction::emitOpenACCCacheConstruct(const OpenACCCacheConstruct &s) { return mlir::success(); } +const VarDecl *getLValueDecl(const Expr *e) { + // We are going to assume that after stripping implicit casts, that the LValue + // is just a DRE around the var-decl. + + e = e->IgnoreImpCasts(); + + const auto *dre = cast<DeclRefExpr>(e); + return cast<VarDecl>(dre->getDecl()); +} + mlir::LogicalResult CIRGenFunction::emitOpenACCAtomicConstruct(const OpenACCAtomicConstruct &s) { - // For now, we are only support 'read'/'write', so diagnose. We can switch on - // the kind later once we start implementing the other 2 forms. While we - if (s.getAtomicKind() != OpenACCAtomicKind::Read && - s.getAtomicKind() != OpenACCAtomicKind::Write) { + // For now, we are only support 'read'/'write'/'update', so diagnose. We can + // switch on the kind later once we implement the 'capture' form. + if (s.getAtomicKind() == OpenACCAtomicKind::Capture) { cgm.errorNYI(s.getSourceRange(), "OpenACC Atomic Construct"); return mlir::failure(); } @@ -318,11 +327,10 @@ CIRGenFunction::emitOpenACCAtomicConstruct(const OpenACCAtomicConstruct &s) { // expression it is associated with rather than emitting it inside of it. So // it has custom emit logic. mlir::Location start = getLoc(s.getSourceRange().getBegin()); + mlir::Location end = getLoc(s.getSourceRange().getEnd()); OpenACCAtomicConstruct::StmtInfo inf = s.getAssociatedStmtInfo(); switch (s.getAtomicKind()) { - case OpenACCAtomicKind::None: - case OpenACCAtomicKind::Update: case OpenACCAtomicKind::Capture: llvm_unreachable("Unimplemented atomic construct type, should have " "diagnosed/returned above"); @@ -353,6 +361,50 @@ CIRGenFunction::emitOpenACCAtomicConstruct(const OpenACCAtomicConstruct &s) { s.clauses()); return mlir::success(); } + case OpenACCAtomicKind::None: + case OpenACCAtomicKind::Update: { + mlir::Value x = emitLValue(inf.X).getPointer(); + auto op = + mlir::acc::AtomicUpdateOp::create(builder, start, x, /*ifCond=*/{}); + emitOpenACCClauses(op, s.getDirectiveKind(), s.getDirectiveLoc(), + s.clauses()); + mlir::LogicalResult res = mlir::success(); + { + mlir::OpBuilder::InsertionGuard guardCase(builder); + mlir::Type argTy = cast<cir::PointerType>(x.getType()).getPointee(); + std::array<mlir::Type, 1> recipeType{argTy}; + std::array<mlir::Location, 1> recipeLoc{start}; + mlir::Block *recipeBlock = builder.createBlock( + &op.getRegion(), op.getRegion().end(), recipeType, recipeLoc); + builder.setInsertionPointToEnd(recipeBlock); + + // Since we have an initial value that we know is a scalar type, we can + // just emit the entire statement here after sneaking-in our 'alloca' in + // the right place, then loading out of it. Flang does a lot less work + // (probably does its own emitting!), but we have more complicated AST + // nodes to worry about, so we can just count on opt to remove the extra + // alloca/load/store set. + auto alloca = cir::AllocaOp::create( + builder, start, x.getType(), argTy, "x_var", + cgm.getSize(getContext().getTypeAlignInChars(inf.X->getType()))); + + alloca.setInitAttr(mlir::UnitAttr::get(&getMLIRContext())); + builder.CIRBaseBuilderTy::createStore(start, recipeBlock->getArgument(0), + alloca); + + const VarDecl *xval = getLValueDecl(inf.X); + CIRGenFunction::DeclMapRevertingRAII declMapRAII{*this, xval}; + replaceAddrOfLocalVar( + xval, Address{alloca, argTy, getContext().getDeclAlign(xval)}); + + res = emitStmt(s.getAssociatedStmt(), /*useCurrentScope=*/true); + + auto load = cir::LoadOp::create(builder, start, {alloca}); + mlir::acc::YieldOp::create(builder, end, {load}); + } + + return res; + } } llvm_unreachable("unknown OpenACC atomic kind"); diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp index 3746bc04..0fea57b 100644 --- a/clang/lib/CodeGen/CodeGenModule.cpp +++ b/clang/lib/CodeGen/CodeGenModule.cpp @@ -146,8 +146,6 @@ createTargetCodeGenInfo(CodeGenModule &CGM) { return createWindowsAArch64TargetCodeGenInfo(CGM, AArch64ABIKind::Win64); else if (Target.getABI() == "aapcs-soft") Kind = AArch64ABIKind::AAPCSSoft; - else if (Target.getABI() == "pauthtest") - Kind = AArch64ABIKind::PAuthTest; return createAArch64TargetCodeGenInfo(CGM, Kind); } diff --git a/clang/lib/CodeGen/TargetBuiltins/RISCV.cpp b/clang/lib/CodeGen/TargetBuiltins/RISCV.cpp index 920d285..1300722 100644 --- a/clang/lib/CodeGen/TargetBuiltins/RISCV.cpp +++ b/clang/lib/CodeGen/TargetBuiltins/RISCV.cpp @@ -1121,6 +1121,8 @@ Value *CodeGenFunction::EmitRISCVBuiltinExpr(unsigned BuiltinID, bool IsMasked = false; // This is used by segment load/store to determine it's llvm type. unsigned SegInstSEW = 8; + // This is used by XSfmm. + unsigned TWiden = 0; // Required for overloaded intrinsics. llvm::SmallVector<llvm::Type *, 2> IntrinsicTypes; diff --git a/clang/lib/CodeGen/TargetInfo.h b/clang/lib/CodeGen/TargetInfo.h index d0edae1..f63e900 100644 --- a/clang/lib/CodeGen/TargetInfo.h +++ b/clang/lib/CodeGen/TargetInfo.h @@ -483,7 +483,6 @@ enum class AArch64ABIKind { DarwinPCS, Win64, AAPCSSoft, - PAuthTest, }; std::unique_ptr<TargetCodeGenInfo> diff --git a/clang/lib/Driver/ToolChain.cpp b/clang/lib/Driver/ToolChain.cpp index 3d5cac6..eea5c2f 100644 --- a/clang/lib/Driver/ToolChain.cpp +++ b/clang/lib/Driver/ToolChain.cpp @@ -1253,7 +1253,6 @@ std::string ToolChain::ComputeLLVMTriple(const ArgList &Args, } case llvm::Triple::aarch64: { llvm::Triple Triple = getTriple(); - tools::aarch64::setPAuthABIInTriple(getDriver(), Args, Triple); if (!Triple.isOSBinFormatMachO()) return Triple.getTriple(); diff --git a/clang/lib/Driver/ToolChains/Arch/AArch64.cpp b/clang/lib/Driver/ToolChains/Arch/AArch64.cpp index eb5d542..e8d5e38 100644 --- a/clang/lib/Driver/ToolChains/Arch/AArch64.cpp +++ b/clang/lib/Driver/ToolChains/Arch/AArch64.cpp @@ -466,27 +466,6 @@ void aarch64::getAArch64TargetFeatures(const Driver &D, Features.push_back("+no-bti-at-return-twice"); } -void aarch64::setPAuthABIInTriple(const Driver &D, const ArgList &Args, - llvm::Triple &Triple) { - Arg *ABIArg = Args.getLastArg(options::OPT_mabi_EQ); - bool HasPAuthABI = - ABIArg ? (StringRef(ABIArg->getValue()) == "pauthtest") : false; - - switch (Triple.getEnvironment()) { - case llvm::Triple::UnknownEnvironment: - if (HasPAuthABI) - Triple.setEnvironment(llvm::Triple::PAuthTest); - break; - case llvm::Triple::PAuthTest: - break; - default: - if (HasPAuthABI) - D.Diag(diag::err_drv_unsupported_opt_for_target) - << ABIArg->getAsString(Args) << Triple.getTriple(); - break; - } -} - /// Is the triple {aarch64.aarch64_be}-none-elf? bool aarch64::isAArch64BareMetal(const llvm::Triple &Triple) { if (Triple.getArch() != llvm::Triple::aarch64 && diff --git a/clang/lib/Driver/ToolChains/Arch/AArch64.h b/clang/lib/Driver/ToolChains/Arch/AArch64.h index 2765ee8..97ebfa6 100644 --- a/clang/lib/Driver/ToolChains/Arch/AArch64.h +++ b/clang/lib/Driver/ToolChains/Arch/AArch64.h @@ -28,8 +28,6 @@ void getAArch64TargetFeatures(const Driver &D, const llvm::Triple &Triple, std::string getAArch64TargetCPU(const llvm::opt::ArgList &Args, const llvm::Triple &Triple, llvm::opt::Arg *&A); -void setPAuthABIInTriple(const Driver &D, const llvm::opt::ArgList &Args, - llvm::Triple &triple); bool isAArch64BareMetal(const llvm::Triple &Triple); } // end namespace aarch64 diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp index caf7478..79edc56 100644 --- a/clang/lib/Driver/ToolChains/Clang.cpp +++ b/clang/lib/Driver/ToolChains/Clang.cpp @@ -1348,59 +1348,6 @@ void AddUnalignedAccessWarning(ArgStringList &CmdArgs) { } } -// Each combination of options here forms a signing schema, and in most cases -// each signing schema is its own incompatible ABI. The default values of the -// options represent the default signing schema. -static void handlePAuthABI(const ArgList &DriverArgs, ArgStringList &CC1Args) { - if (!DriverArgs.hasArg(options::OPT_fptrauth_intrinsics, - options::OPT_fno_ptrauth_intrinsics)) - CC1Args.push_back("-fptrauth-intrinsics"); - - if (!DriverArgs.hasArg(options::OPT_fptrauth_calls, - options::OPT_fno_ptrauth_calls)) - CC1Args.push_back("-fptrauth-calls"); - - if (!DriverArgs.hasArg(options::OPT_fptrauth_returns, - options::OPT_fno_ptrauth_returns)) - CC1Args.push_back("-fptrauth-returns"); - - if (!DriverArgs.hasArg(options::OPT_fptrauth_auth_traps, - options::OPT_fno_ptrauth_auth_traps)) - CC1Args.push_back("-fptrauth-auth-traps"); - - if (!DriverArgs.hasArg( - options::OPT_fptrauth_vtable_pointer_address_discrimination, - options::OPT_fno_ptrauth_vtable_pointer_address_discrimination)) - CC1Args.push_back("-fptrauth-vtable-pointer-address-discrimination"); - - if (!DriverArgs.hasArg( - options::OPT_fptrauth_vtable_pointer_type_discrimination, - options::OPT_fno_ptrauth_vtable_pointer_type_discrimination)) - CC1Args.push_back("-fptrauth-vtable-pointer-type-discrimination"); - - if (!DriverArgs.hasArg( - options::OPT_fptrauth_type_info_vtable_pointer_discrimination, - options::OPT_fno_ptrauth_type_info_vtable_pointer_discrimination)) - CC1Args.push_back("-fptrauth-type-info-vtable-pointer-discrimination"); - - if (!DriverArgs.hasArg(options::OPT_fptrauth_indirect_gotos, - options::OPT_fno_ptrauth_indirect_gotos)) - CC1Args.push_back("-fptrauth-indirect-gotos"); - - if (!DriverArgs.hasArg(options::OPT_fptrauth_init_fini, - options::OPT_fno_ptrauth_init_fini)) - CC1Args.push_back("-fptrauth-init-fini"); - - if (!DriverArgs.hasArg( - options::OPT_fptrauth_init_fini_address_discrimination, - options::OPT_fno_ptrauth_init_fini_address_discrimination)) - CC1Args.push_back("-fptrauth-init-fini-address-discrimination"); - - if (!DriverArgs.hasArg(options::OPT_faarch64_jump_table_hardening, - options::OPT_fno_aarch64_jump_table_hardening)) - CC1Args.push_back("-faarch64-jump-table-hardening"); -} - static void CollectARMPACBTIOptions(const ToolChain &TC, const ArgList &Args, ArgStringList &CmdArgs, bool isAArch64) { const llvm::Triple &Triple = TC.getEffectiveTriple(); @@ -1638,7 +1585,9 @@ void RenderAArch64ABI(const llvm::Triple &Triple, const ArgList &Args, ABIName = A->getValue(); else if (Triple.isOSDarwin()) ABIName = "darwinpcs"; - else if (Triple.getEnvironment() == llvm::Triple::PAuthTest) + // TODO: we probably want to have some target hook here. + else if (Triple.isOSLinux() && + Triple.getEnvironment() == llvm::Triple::PAuthTest) ABIName = "pauthtest"; else ABIName = "aapcs"; @@ -1758,8 +1707,6 @@ void Clang::AddAArch64TargetArgs(const ArgList &Args, options::OPT_fno_ptrauth_objc_interface_sel); Args.addOptInFlag(CmdArgs, options::OPT_fptrauth_objc_class_ro, options::OPT_fno_ptrauth_objc_class_ro); - if (Triple.getEnvironment() == llvm::Triple::PAuthTest) - handlePAuthABI(Args, CmdArgs); // Enable/disable return address signing and indirect branch targets. CollectARMPACBTIOptions(getToolChain(), Args, CmdArgs, true /*isAArch64*/); diff --git a/clang/lib/Driver/ToolChains/Fuchsia.cpp b/clang/lib/Driver/ToolChains/Fuchsia.cpp index 146dc8b..31c2f3f 100644 --- a/clang/lib/Driver/ToolChains/Fuchsia.cpp +++ b/clang/lib/Driver/ToolChains/Fuchsia.cpp @@ -481,9 +481,11 @@ SanitizerMask Fuchsia::getSupportedSanitizers() const { Res |= SanitizerKind::Fuzzer; Res |= SanitizerKind::FuzzerNoLink; Res |= SanitizerKind::Leak; - Res |= SanitizerKind::SafeStack; Res |= SanitizerKind::Scudo; Res |= SanitizerKind::Thread; + if (getTriple().getArch() == llvm::Triple::x86_64) { + Res |= SanitizerKind::SafeStack; + } return Res; } diff --git a/clang/lib/Driver/ToolChains/Linux.cpp b/clang/lib/Driver/ToolChains/Linux.cpp index 8eb4d34e..94a9fe8 100644 --- a/clang/lib/Driver/ToolChains/Linux.cpp +++ b/clang/lib/Driver/ToolChains/Linux.cpp @@ -445,6 +445,102 @@ std::string Linux::computeSysRoot() const { return std::string(); } +static void setPAuthABIInTriple(const Driver &D, const ArgList &Args, + llvm::Triple &Triple) { + Arg *ABIArg = Args.getLastArg(options::OPT_mabi_EQ); + bool HasPAuthABI = + ABIArg ? (StringRef(ABIArg->getValue()) == "pauthtest") : false; + + switch (Triple.getEnvironment()) { + case llvm::Triple::UnknownEnvironment: + if (HasPAuthABI) + Triple.setEnvironment(llvm::Triple::PAuthTest); + break; + case llvm::Triple::PAuthTest: + break; + default: + if (HasPAuthABI) + D.Diag(diag::err_drv_unsupported_opt_for_target) + << ABIArg->getAsString(Args) << Triple.getTriple(); + break; + } +} + +std::string Linux::ComputeEffectiveClangTriple(const llvm::opt::ArgList &Args, + types::ID InputType) const { + std::string TripleString = + Generic_ELF::ComputeEffectiveClangTriple(Args, InputType); + if (getTriple().isAArch64()) { + llvm::Triple Triple(TripleString); + setPAuthABIInTriple(getDriver(), Args, Triple); + return Triple.getTriple(); + } + return TripleString; +} + +// Each combination of options here forms a signing schema, and in most cases +// each signing schema is its own incompatible ABI. The default values of the +// options represent the default signing schema. +static void handlePAuthABI(const Driver &D, const ArgList &DriverArgs, + ArgStringList &CC1Args) { + if (!DriverArgs.hasArg(options::OPT_fptrauth_intrinsics, + options::OPT_fno_ptrauth_intrinsics)) + CC1Args.push_back("-fptrauth-intrinsics"); + + if (!DriverArgs.hasArg(options::OPT_fptrauth_calls, + options::OPT_fno_ptrauth_calls)) + CC1Args.push_back("-fptrauth-calls"); + + if (!DriverArgs.hasArg(options::OPT_fptrauth_returns, + options::OPT_fno_ptrauth_returns)) + CC1Args.push_back("-fptrauth-returns"); + + if (!DriverArgs.hasArg(options::OPT_fptrauth_auth_traps, + options::OPT_fno_ptrauth_auth_traps)) + CC1Args.push_back("-fptrauth-auth-traps"); + + if (!DriverArgs.hasArg( + options::OPT_fptrauth_vtable_pointer_address_discrimination, + options::OPT_fno_ptrauth_vtable_pointer_address_discrimination)) + CC1Args.push_back("-fptrauth-vtable-pointer-address-discrimination"); + + if (!DriverArgs.hasArg( + options::OPT_fptrauth_vtable_pointer_type_discrimination, + options::OPT_fno_ptrauth_vtable_pointer_type_discrimination)) + CC1Args.push_back("-fptrauth-vtable-pointer-type-discrimination"); + + if (!DriverArgs.hasArg( + options::OPT_fptrauth_type_info_vtable_pointer_discrimination, + options::OPT_fno_ptrauth_type_info_vtable_pointer_discrimination)) + CC1Args.push_back("-fptrauth-type-info-vtable-pointer-discrimination"); + + if (!DriverArgs.hasArg(options::OPT_fptrauth_indirect_gotos, + options::OPT_fno_ptrauth_indirect_gotos)) + CC1Args.push_back("-fptrauth-indirect-gotos"); + + if (!DriverArgs.hasArg(options::OPT_fptrauth_init_fini, + options::OPT_fno_ptrauth_init_fini)) + CC1Args.push_back("-fptrauth-init-fini"); + + if (!DriverArgs.hasArg( + options::OPT_fptrauth_init_fini_address_discrimination, + options::OPT_fno_ptrauth_init_fini_address_discrimination)) + CC1Args.push_back("-fptrauth-init-fini-address-discrimination"); + + if (!DriverArgs.hasArg(options::OPT_faarch64_jump_table_hardening, + options::OPT_fno_aarch64_jump_table_hardening)) + CC1Args.push_back("-faarch64-jump-table-hardening"); +} + +void Linux::addClangTargetOptions(const llvm::opt::ArgList &DriverArgs, + llvm::opt::ArgStringList &CC1Args, + Action::OffloadKind DeviceOffloadKind) const { + llvm::Triple Triple(ComputeEffectiveClangTriple(DriverArgs)); + if (Triple.isAArch64() && Triple.getEnvironment() == llvm::Triple::PAuthTest) + handlePAuthABI(getDriver(), DriverArgs, CC1Args); + Generic_ELF::addClangTargetOptions(DriverArgs, CC1Args, DeviceOffloadKind); +} + std::string Linux::getDynamicLinker(const ArgList &Args) const { const llvm::Triple::ArchType Arch = getArch(); const llvm::Triple &Triple = getTriple(); diff --git a/clang/lib/Driver/ToolChains/Linux.h b/clang/lib/Driver/ToolChains/Linux.h index 2eb2d05..97bad77 100644 --- a/clang/lib/Driver/ToolChains/Linux.h +++ b/clang/lib/Driver/ToolChains/Linux.h @@ -53,7 +53,14 @@ public: SanitizerMask getSupportedSanitizers() const override; void addProfileRTLibs(const llvm::opt::ArgList &Args, llvm::opt::ArgStringList &CmdArgs) const override; + std::string ComputeEffectiveClangTriple( + const llvm::opt::ArgList &Args, + types::ID InputType = types::TY_INVALID) const override; std::string computeSysRoot() const override; + void + addClangTargetOptions(const llvm::opt::ArgList &DriverArgs, + llvm::opt::ArgStringList &CC1Args, + Action::OffloadKind DeviceOffloadKind) const override; std::string getDynamicLinker(const llvm::opt::ArgList &Args) const override; diff --git a/clang/lib/Frontend/CompilerInstance.cpp b/clang/lib/Frontend/CompilerInstance.cpp index e3bf0ea..6b09f7f 100644 --- a/clang/lib/Frontend/CompilerInstance.cpp +++ b/clang/lib/Frontend/CompilerInstance.cpp @@ -882,7 +882,7 @@ CompilerInstance::createOutputFileImpl(StringRef OutputPath, bool Binary, "File Manager is required to fix up relative path.\n"); AbsPath.emplace(OutputPath); - FileMgr->FixupRelativePath(*AbsPath); + FileManager::fixupRelativePath(getFileSystemOpts(), *AbsPath); OutputPath = *AbsPath; } diff --git a/clang/lib/Headers/sifive_vector.h b/clang/lib/Headers/sifive_vector.h index 4e67ad6..ae01627 100644 --- a/clang/lib/Headers/sifive_vector.h +++ b/clang/lib/Headers/sifive_vector.h @@ -115,4 +115,60 @@ #endif #endif +#define __riscv_sf_vsettnt_e8w1(atn) __riscv_sf_vsettnt(atn, 0, 1); +#define __riscv_sf_vsettnt_e8w2(atn) __riscv_sf_vsettnt(atn, 0, 2); +#define __riscv_sf_vsettnt_e8w4(atn) __riscv_sf_vsettnt(atn, 0, 3); +#define __riscv_sf_vsettnt_e16w1(atn) __riscv_sf_vsettnt(atn, 1, 1); +#define __riscv_sf_vsettnt_e16w2(atn) __riscv_sf_vsettnt(atn, 1, 2); +#define __riscv_sf_vsettnt_e16w4(atn) __riscv_sf_vsettnt(atn, 1, 3); +#define __riscv_sf_vsettnt_e32w1(atn) __riscv_sf_vsettnt(atn, 2, 1); +#define __riscv_sf_vsettnt_e32w2(atn) __riscv_sf_vsettnt(atn, 2, 2); +#define __riscv_sf_vsettm_e8w1(atm) __riscv_sf_vsettm(atm, 0, 1); +#define __riscv_sf_vsettm_e8w2(atm) __riscv_sf_vsettm(atm, 0, 2); +#define __riscv_sf_vsettm_e8w4(atm) __riscv_sf_vsettm(atm, 0, 3); +#define __riscv_sf_vsettm_e16w1(atm) __riscv_sf_vsettm(atm, 1, 1); +#define __riscv_sf_vsettm_e16w2(atm) __riscv_sf_vsettm(atm, 1, 2); +#define __riscv_sf_vsettm_e16w4(atm) __riscv_sf_vsettm(atm, 1, 3); +#define __riscv_sf_vsettm_e32w1(atm) __riscv_sf_vsettm(atm, 2, 1); +#define __riscv_sf_vsettm_e32w2(atm) __riscv_sf_vsettm(atm, 2, 2); +#define __riscv_sf_vsettn_e8w1(atn) __riscv_sf_vsettn(atn, 0, 1); +#define __riscv_sf_vsettn_e8w2(atn) __riscv_sf_vsettn(atn, 0, 2); +#define __riscv_sf_vsettn_e8w4(atn) __riscv_sf_vsettn(atn, 0, 3); +#define __riscv_sf_vsettn_e16w1(atn) __riscv_sf_vsettn(atn, 1, 1); +#define __riscv_sf_vsettn_e16w2(atn) __riscv_sf_vsettn(atn, 1, 2); +#define __riscv_sf_vsettn_e16w4(atn) __riscv_sf_vsettn(atn, 1, 3); +#define __riscv_sf_vsettn_e32w1(atn) __riscv_sf_vsettn(atn, 2, 1); +#define __riscv_sf_vsettn_e32w2(atn) __riscv_sf_vsettn(atn, 2, 2); +#define __riscv_sf_vsettk_e8w1(atk) __riscv_sf_vsettk(atk, 0, 1); +#define __riscv_sf_vsettk_e8w2(atk) __riscv_sf_vsettk(atk, 0, 2); +#define __riscv_sf_vsettk_e8w4(atk) __riscv_sf_vsettk(atk, 0, 3); +#define __riscv_sf_vsettk_e16w1(atk) __riscv_sf_vsettk(atk, 1, 1); +#define __riscv_sf_vsettk_e16w2(atk) __riscv_sf_vsettk(atk, 1, 2); +#define __riscv_sf_vsettk_e16w4(atk) __riscv_sf_vsettk(atk, 1, 3); +#define __riscv_sf_vsettk_e32w1(atk) __riscv_sf_vsettk(atk, 2, 1); +#define __riscv_sf_vsettk_e32w2(atk) __riscv_sf_vsettk(atk, 2, 2); +#define __riscv_sf_vtzero_t_e8w1(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 3, 1); +#define __riscv_sf_vtzero_t_e8w2(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 3, 2); +#define __riscv_sf_vtzero_t_e8w4(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 3, 4); +#define __riscv_sf_vtzero_t_e16w1(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 4, 1); +#define __riscv_sf_vtzero_t_e16w2(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 4, 2); +#define __riscv_sf_vtzero_t_e16w4(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 4, 4); +#define __riscv_sf_vtzero_t_e32w1(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 5, 1); +#define __riscv_sf_vtzero_t_e32w2(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 5, 2); +#if __riscv_v_elen >= 64 +#define __riscv_sf_vsettnt_e64w1(atn) __riscv_sf_vsettnt(atn, 3, 1); +#define __riscv_sf_vsettm_e64w1(atm) __riscv_sf_vsettm(atm, 3, 1); +#define __riscv_sf_vsettn_e64w1(atn) __riscv_sf_vsettn(atn, 3, 1); +#define __riscv_sf_vsettk_e64w1(atk) __riscv_sf_vsettk(atk, 3, 1); +#define __riscv_sf_vtzero_t_e64w1(tile, atm, atn) \ + __riscv_sf_vtzero_t(tile, atm, atn, 6, 1); +#endif #endif //_SIFIVE_VECTOR_H_ diff --git a/clang/lib/Parse/ParseHLSL.cpp b/clang/lib/Parse/ParseHLSL.cpp index 51f2aef..c727ee3 100644 --- a/clang/lib/Parse/ParseHLSL.cpp +++ b/clang/lib/Parse/ParseHLSL.cpp @@ -126,15 +126,9 @@ Parser::ParsedSemantic Parser::ParseHLSLSemantic() { // semantic index. The semantic index is the number at the end of // the semantic, including leading zeroes. Digits located before // the last letter are part of the semantic name. - bool Invalid = false; SmallString<256> Buffer; Buffer.resize(Tok.getLength() + 1); StringRef Identifier = PP.getSpelling(Tok, Buffer); - if (Invalid) { - Diag(Tok.getLocation(), diag::err_expected_semantic_identifier); - return {}; - } - assert(Identifier.size() > 0); // Determine the start of the semantic index. unsigned IndexIndex = Identifier.find_last_not_of("0123456789") + 1; diff --git a/clang/lib/Sema/CheckExprLifetime.cpp b/clang/lib/Sema/CheckExprLifetime.cpp index e797400..f9665b5 100644 --- a/clang/lib/Sema/CheckExprLifetime.cpp +++ b/clang/lib/Sema/CheckExprLifetime.cpp @@ -155,6 +155,7 @@ getEntityLifetime(const InitializedEntity *Entity, case InitializedEntity::EK_LambdaToBlockConversionBlockElement: case InitializedEntity::EK_LambdaCapture: case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: return {nullptr, LK_FullExpression}; diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp index f99c01e..f451787 100644 --- a/clang/lib/Sema/SemaChecking.cpp +++ b/clang/lib/Sema/SemaChecking.cpp @@ -12373,14 +12373,9 @@ static void DiagnoseMixedUnicodeImplicitConversion(Sema &S, const Type *Source, } } -enum CFIUncheckedCalleeChange { - None, - Adding, - Discarding, -}; - -static CFIUncheckedCalleeChange AdjustingCFIUncheckedCallee(QualType From, - QualType To) { +bool Sema::DiscardingCFIUncheckedCallee(QualType From, QualType To) const { + From = Context.getCanonicalType(From); + To = Context.getCanonicalType(To); QualType MaybePointee = From->getPointeeType(); if (!MaybePointee.isNull() && MaybePointee->getAs<FunctionType>()) From = MaybePointee; @@ -12392,25 +12387,10 @@ static CFIUncheckedCalleeChange AdjustingCFIUncheckedCallee(QualType From, if (const auto *ToFn = To->getAs<FunctionType>()) { if (FromFn->getCFIUncheckedCalleeAttr() && !ToFn->getCFIUncheckedCalleeAttr()) - return Discarding; - if (!FromFn->getCFIUncheckedCalleeAttr() && - ToFn->getCFIUncheckedCalleeAttr()) - return Adding; + return true; } } - return None; -} - -bool Sema::DiscardingCFIUncheckedCallee(QualType From, QualType To) const { - From = Context.getCanonicalType(From); - To = Context.getCanonicalType(To); - return ::AdjustingCFIUncheckedCallee(From, To) == Discarding; -} - -bool Sema::AddingCFIUncheckedCallee(QualType From, QualType To) const { - From = Context.getCanonicalType(From); - To = Context.getCanonicalType(To); - return ::AdjustingCFIUncheckedCallee(From, To) == Adding; + return false; } void Sema::CheckImplicitConversion(Expr *E, QualType T, SourceLocation CC, diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp index 2a485da..96d5142 100644 --- a/clang/lib/Sema/SemaHLSL.cpp +++ b/clang/lib/Sema/SemaHLSL.cpp @@ -21,6 +21,7 @@ #include "clang/AST/Expr.h" #include "clang/AST/HLSLResource.h" #include "clang/AST/Type.h" +#include "clang/AST/TypeBase.h" #include "clang/AST/TypeLoc.h" #include "clang/Basic/Builtins.h" #include "clang/Basic/DiagnosticSema.h" @@ -3432,6 +3433,11 @@ static void BuildFlattenedTypeList(QualType BaseTy, List.insert(List.end(), VT->getNumElements(), VT->getElementType()); continue; } + if (const auto *MT = dyn_cast<ConstantMatrixType>(T)) { + List.insert(List.end(), MT->getNumElementsFlattened(), + MT->getElementType()); + continue; + } if (const auto *RD = T->getAsCXXRecordDecl()) { if (RD->isStandardLayout()) RD = RD->getStandardLayoutBaseWithFields(); @@ -4230,6 +4236,32 @@ class InitListTransformer { } return true; } + if (auto *MTy = Ty->getAs<ConstantMatrixType>()) { + unsigned Rows = MTy->getNumRows(); + unsigned Cols = MTy->getNumColumns(); + QualType ElemTy = MTy->getElementType(); + + for (unsigned C = 0; C < Cols; ++C) { + for (unsigned R = 0; R < Rows; ++R) { + // row index literal + Expr *RowIdx = IntegerLiteral::Create( + Ctx, llvm::APInt(Ctx.getIntWidth(Ctx.IntTy), R), Ctx.IntTy, + E->getBeginLoc()); + // column index literal + Expr *ColIdx = IntegerLiteral::Create( + Ctx, llvm::APInt(Ctx.getIntWidth(Ctx.IntTy), C), Ctx.IntTy, + E->getBeginLoc()); + ExprResult ElExpr = S.CreateBuiltinMatrixSubscriptExpr( + E, RowIdx, ColIdx, E->getEndLoc()); + if (ElExpr.isInvalid()) + return false; + if (!castInitializer(ElExpr.get())) + return false; + ElExpr.get()->setType(ElemTy); + } + } + return true; + } if (auto *ArrTy = dyn_cast<ConstantArrayType>(Ty.getTypePtr())) { uint64_t Size = ArrTy->getZExtSize(); @@ -4283,14 +4315,17 @@ class InitListTransformer { return *(ArgIt++); llvm::SmallVector<Expr *> Inits; - assert(!isa<MatrixType>(Ty) && "Matrix types not yet supported in HLSL"); Ty = Ty.getDesugaredType(Ctx); - if (Ty->isVectorType() || Ty->isConstantArrayType()) { + if (Ty->isVectorType() || Ty->isConstantArrayType() || + Ty->isConstantMatrixType()) { QualType ElTy; uint64_t Size = 0; if (auto *ATy = Ty->getAs<VectorType>()) { ElTy = ATy->getElementType(); Size = ATy->getNumElements(); + } else if (auto *CMTy = Ty->getAs<ConstantMatrixType>()) { + ElTy = CMTy->getElementType(); + Size = CMTy->getNumElementsFlattened(); } else { auto *VTy = cast<ConstantArrayType>(Ty.getTypePtr()); ElTy = VTy->getElementType(); diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp index 7debe33..073010d 100644 --- a/clang/lib/Sema/SemaInit.cpp +++ b/clang/lib/Sema/SemaInit.cpp @@ -17,6 +17,7 @@ #include "clang/AST/ExprCXX.h" #include "clang/AST/ExprObjC.h" #include "clang/AST/IgnoreExpr.h" +#include "clang/AST/TypeBase.h" #include "clang/AST/TypeLoc.h" #include "clang/Basic/SourceManager.h" #include "clang/Basic/Specifiers.h" @@ -403,6 +404,9 @@ class InitListChecker { unsigned &Index, InitListExpr *StructuredList, unsigned &StructuredIndex); + void CheckMatrixType(const InitializedEntity &Entity, InitListExpr *IList, + QualType DeclType, unsigned &Index, + InitListExpr *StructuredList, unsigned &StructuredIndex); void CheckVectorType(const InitializedEntity &Entity, InitListExpr *IList, QualType DeclType, unsigned &Index, InitListExpr *StructuredList, @@ -1004,7 +1008,8 @@ InitListChecker::FillInEmptyInitializations(const InitializedEntity &Entity, return; if (ElementEntity.getKind() == InitializedEntity::EK_ArrayElement || - ElementEntity.getKind() == InitializedEntity::EK_VectorElement) + ElementEntity.getKind() == InitializedEntity::EK_VectorElement || + ElementEntity.getKind() == InitializedEntity::EK_MatrixElement) ElementEntity.setElementIndex(Init); if (Init >= NumInits && (ILE->hasArrayFiller() || SkipEmptyInitChecks)) @@ -1274,6 +1279,7 @@ static void warnBracedScalarInit(Sema &S, const InitializedEntity &Entity, switch (Entity.getKind()) { case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: case InitializedEntity::EK_ArrayElement: case InitializedEntity::EK_Parameter: @@ -1373,11 +1379,12 @@ void InitListChecker::CheckExplicitInitList(const InitializedEntity &Entity, SemaRef.Diag(IList->getInit(Index)->getBeginLoc(), DK) << T << IList->getInit(Index)->getSourceRange(); } else { - int initKind = T->isArrayType() ? 0 : - T->isVectorType() ? 1 : - T->isScalarType() ? 2 : - T->isUnionType() ? 3 : - 4; + int initKind = T->isArrayType() ? 0 + : T->isVectorType() ? 1 + : T->isMatrixType() ? 2 + : T->isScalarType() ? 3 + : T->isUnionType() ? 4 + : 5; unsigned DK = ExtraInitsIsError ? diag::err_excess_initializers : diag::ext_excess_initializers; @@ -1431,6 +1438,9 @@ void InitListChecker::CheckListElementTypes(const InitializedEntity &Entity, } else if (DeclType->isVectorType()) { CheckVectorType(Entity, IList, DeclType, Index, StructuredList, StructuredIndex); + } else if (DeclType->isMatrixType()) { + CheckMatrixType(Entity, IList, DeclType, Index, StructuredList, + StructuredIndex); } else if (const RecordDecl *RD = DeclType->getAsRecordDecl()) { auto Bases = CXXRecordDecl::base_class_const_range(CXXRecordDecl::base_class_const_iterator(), @@ -1878,6 +1888,37 @@ void InitListChecker::CheckReferenceType(const InitializedEntity &Entity, AggrDeductionCandidateParamTypes->push_back(DeclType); } +void InitListChecker::CheckMatrixType(const InitializedEntity &Entity, + InitListExpr *IList, QualType DeclType, + unsigned &Index, + InitListExpr *StructuredList, + unsigned &StructuredIndex) { + if (!SemaRef.getLangOpts().HLSL) + return; + + const ConstantMatrixType *MT = DeclType->castAs<ConstantMatrixType>(); + QualType ElemTy = MT->getElementType(); + const unsigned MaxElts = MT->getNumElementsFlattened(); + + unsigned NumEltsInit = 0; + InitializedEntity ElemEnt = + InitializedEntity::InitializeElement(SemaRef.Context, 0, Entity); + + while (NumEltsInit < MaxElts && Index < IList->getNumInits()) { + // Not a sublist: just consume directly. + ElemEnt.setElementIndex(Index); + CheckSubElementType(ElemEnt, IList, ElemTy, Index, StructuredList, + StructuredIndex); + ++NumEltsInit; + } + + // For HLSL The error for this case is handled in SemaHLSL's initializer + // list diagnostics, That means the execution should require NumEltsInit + // to equal Max initializers. In other words execution should never + // reach this point if this condition is not true". + assert(NumEltsInit == MaxElts && "NumEltsInit must equal MaxElts"); +} + void InitListChecker::CheckVectorType(const InitializedEntity &Entity, InitListExpr *IList, QualType DeclType, unsigned &Index, @@ -3640,6 +3681,9 @@ InitializedEntity::InitializedEntity(ASTContext &Context, unsigned Index, } else if (const VectorType *VT = Parent.getType()->getAs<VectorType>()) { Kind = EK_VectorElement; Type = VT->getElementType(); + } else if (const MatrixType *MT = Parent.getType()->getAs<MatrixType>()) { + Kind = EK_MatrixElement; + Type = MT->getElementType(); } else { const ComplexType *CT = Parent.getType()->getAs<ComplexType>(); assert(CT && "Unexpected type"); @@ -3688,6 +3732,7 @@ DeclarationName InitializedEntity::getName() const { case EK_Delegating: case EK_ArrayElement: case EK_VectorElement: + case EK_MatrixElement: case EK_ComplexElement: case EK_BlockElement: case EK_LambdaToBlockConversionBlockElement: @@ -3721,6 +3766,7 @@ ValueDecl *InitializedEntity::getDecl() const { case EK_Delegating: case EK_ArrayElement: case EK_VectorElement: + case EK_MatrixElement: case EK_ComplexElement: case EK_BlockElement: case EK_LambdaToBlockConversionBlockElement: @@ -3754,6 +3800,7 @@ bool InitializedEntity::allowsNRVO() const { case EK_Delegating: case EK_ArrayElement: case EK_VectorElement: + case EK_MatrixElement: case EK_ComplexElement: case EK_BlockElement: case EK_LambdaToBlockConversionBlockElement: @@ -3793,6 +3840,9 @@ unsigned InitializedEntity::dumpImpl(raw_ostream &OS) const { case EK_Delegating: OS << "Delegating"; break; case EK_ArrayElement: OS << "ArrayElement " << Index; break; case EK_VectorElement: OS << "VectorElement " << Index; break; + case EK_MatrixElement: + OS << "MatrixElement " << Index; + break; case EK_ComplexElement: OS << "ComplexElement " << Index; break; case EK_BlockElement: OS << "Block"; break; case EK_LambdaToBlockConversionBlockElement: @@ -6030,7 +6080,7 @@ static void TryOrBuildParenListInitialization( Sequence.SetFailed(InitializationSequence::FK_ParenthesizedListInitFailed); if (!VerifyOnly) { QualType T = Entity.getType(); - int InitKind = T->isArrayType() ? 0 : T->isUnionType() ? 3 : 4; + int InitKind = T->isArrayType() ? 0 : T->isUnionType() ? 4 : 5; SourceRange ExcessInitSR(Args[EntityIndexToProcess]->getBeginLoc(), Args.back()->getEndLoc()); S.Diag(Kind.getLocation(), diag::err_excess_initializers) @@ -6823,7 +6873,8 @@ void InitializationSequence::InitializeFrom(Sema &S, // For HLSL ext vector types we allow list initialization behavior for C++ // functional cast expressions which look like constructor syntax. This is // accomplished by converting initialization arguments to InitListExpr. - if (S.getLangOpts().HLSL && Args.size() > 1 && DestType->isExtVectorType() && + if (S.getLangOpts().HLSL && Args.size() > 1 && + (DestType->isExtVectorType() || DestType->isConstantMatrixType()) && (SourceType.isNull() || !Context.hasSameUnqualifiedType(SourceType, DestType))) { InitListExpr *ILE = new (Context) @@ -6988,6 +7039,7 @@ static AssignmentAction getAssignmentAction(const InitializedEntity &Entity, case InitializedEntity::EK_Binding: case InitializedEntity::EK_ArrayElement: case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: case InitializedEntity::EK_BlockElement: case InitializedEntity::EK_LambdaToBlockConversionBlockElement: @@ -7013,6 +7065,7 @@ static bool shouldBindAsTemporary(const InitializedEntity &Entity) { case InitializedEntity::EK_Base: case InitializedEntity::EK_Delegating: case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: case InitializedEntity::EK_Exception: case InitializedEntity::EK_BlockElement: @@ -7043,6 +7096,7 @@ static bool shouldDestroyEntity(const InitializedEntity &Entity) { case InitializedEntity::EK_Base: case InitializedEntity::EK_Delegating: case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: case InitializedEntity::EK_BlockElement: case InitializedEntity::EK_LambdaToBlockConversionBlockElement: @@ -7096,6 +7150,7 @@ static SourceLocation getInitializationLoc(const InitializedEntity &Entity, case InitializedEntity::EK_Base: case InitializedEntity::EK_Delegating: case InitializedEntity::EK_VectorElement: + case InitializedEntity::EK_MatrixElement: case InitializedEntity::EK_ComplexElement: case InitializedEntity::EK_BlockElement: case InitializedEntity::EK_LambdaToBlockConversionBlockElement: @@ -7845,11 +7900,13 @@ ExprResult InitializationSequence::Perform(Sema &S, ExprResult CurInit((Expr *)nullptr); SmallVector<Expr*, 4> ArrayLoopCommonExprs; - // HLSL allows vector initialization to function like list initialization, but - // use the syntax of a C++-like constructor. - bool IsHLSLVectorInit = S.getLangOpts().HLSL && DestType->isExtVectorType() && - isa<InitListExpr>(Args[0]); - (void)IsHLSLVectorInit; + // HLSL allows vector/matrix initialization to function like list + // initialization, but use the syntax of a C++-like constructor. + bool IsHLSLVectorOrMatrixInit = + S.getLangOpts().HLSL && + (DestType->isExtVectorType() || DestType->isConstantMatrixType()) && + isa<InitListExpr>(Args[0]); + (void)IsHLSLVectorOrMatrixInit; // For initialization steps that start with a single initializer, // grab the only argument out the Args and place it into the "current" @@ -7888,7 +7945,7 @@ ExprResult InitializationSequence::Perform(Sema &S, case SK_StdInitializerList: case SK_OCLSamplerInit: case SK_OCLZeroOpaqueType: { - assert(Args.size() == 1 || IsHLSLVectorInit); + assert(Args.size() == 1 || IsHLSLVectorOrMatrixInit); CurInit = Args[0]; if (!CurInit.get()) return ExprError(); break; @@ -9105,7 +9162,7 @@ bool InitializationSequence::Diagnose(Sema &S, << R; else S.Diag(Kind.getLocation(), diag::err_excess_initializers) - << /*scalar=*/2 << R; + << /*scalar=*/3 << R; break; } diff --git a/clang/lib/Sema/SemaObjCProperty.cpp b/clang/lib/Sema/SemaObjCProperty.cpp index 1880cec..67c554c 100644 --- a/clang/lib/Sema/SemaObjCProperty.cpp +++ b/clang/lib/Sema/SemaObjCProperty.cpp @@ -1041,7 +1041,7 @@ RedeclarePropertyAccessor(ASTContext &Context, ObjCImplementationDecl *Impl, Decl->getSelector(), Decl->getReturnType(), Decl->getReturnTypeSourceInfo(), Impl, Decl->isInstanceMethod(), Decl->isVariadic(), Decl->isPropertyAccessor(), - /* isSynthesized*/ true, Decl->isImplicit(), Decl->isDefined(), + /*isSynthesizedAccessorStub=*/true, Decl->isImplicit(), Decl->isDefined(), Decl->getImplementationControl(), Decl->hasRelatedResultType()); ImplDecl->getMethodFamily(); if (Decl->hasAttrs()) diff --git a/clang/lib/Sema/SemaOverload.cpp b/clang/lib/Sema/SemaOverload.cpp index 1f25111..37f3511 100644 --- a/clang/lib/Sema/SemaOverload.cpp +++ b/clang/lib/Sema/SemaOverload.cpp @@ -2532,15 +2532,12 @@ static bool IsStandardConversion(Sema &S, Expr* From, QualType ToType, SCS.setToType(2, FromType); - // If we have not converted the argument type to the parameter type, - // this is a bad conversion sequence, unless we're resolving an overload in C. - // - // Permit conversions from a function without `cfi_unchecked_callee` to a - // function with `cfi_unchecked_callee`. - if (CanonFrom == CanonTo || S.AddingCFIUncheckedCallee(CanonFrom, CanonTo)) + if (CanonFrom == CanonTo) return true; - if ((S.getLangOpts().CPlusPlus || !InOverloadResolution)) + // If we have not converted the argument type to the parameter type, + // this is a bad conversion sequence, unless we're resolving an overload in C. + if (S.getLangOpts().CPlusPlus || !InOverloadResolution) return false; ExprResult ER = ExprResult{From}; diff --git a/clang/lib/Sema/SemaRISCV.cpp b/clang/lib/Sema/SemaRISCV.cpp index b5f91a3..75dba80 100644 --- a/clang/lib/Sema/SemaRISCV.cpp +++ b/clang/lib/Sema/SemaRISCV.cpp @@ -664,6 +664,80 @@ bool SemaRISCV::CheckBuiltinFunctionCall(const TargetInfo &TI, return CheckVSetVL(1, 2); case RISCVVector::BI__builtin_rvv_vsetvlimax: return CheckVSetVL(0, 1); + case RISCVVector::BI__builtin_rvv_sf_vsettnt: + case RISCVVector::BI__builtin_rvv_sf_vsettm: + case RISCVVector::BI__builtin_rvv_sf_vsettn: + case RISCVVector::BI__builtin_rvv_sf_vsettk: + return SemaRef.BuiltinConstantArgRange(TheCall, 1, 0, 3) || + SemaRef.BuiltinConstantArgRange(TheCall, 2, 1, 3); + case RISCVVector::BI__builtin_rvv_sf_mm_f_f_w1: + case RISCVVector::BI__builtin_rvv_sf_mm_f_f_w2: + case RISCVVector::BI__builtin_rvv_sf_mm_e5m2_e4m3_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_e5m2_e5m2_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_e4m3_e4m3_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_e4m3_e5m2_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_u_u_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_u_s_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_s_u_w4: + case RISCVVector::BI__builtin_rvv_sf_mm_s_s_w4: { + QualType Arg1Type = TheCall->getArg(1)->getType(); + ASTContext::BuiltinVectorTypeInfo Info = + SemaRef.Context.getBuiltinVectorTypeInfo( + Arg1Type->castAs<BuiltinType>()); + unsigned EltSize = SemaRef.Context.getTypeSize(Info.ElementType); + llvm::APSInt Result; + + // We can't check the value of a dependent argument. + Expr *Arg = TheCall->getArg(0); + if (Arg->isTypeDependent() || Arg->isValueDependent()) + return false; + + // Check constant-ness first. + if (SemaRef.BuiltinConstantArg(TheCall, 0, Result)) + return true; + + // For TEW = 32, mtd can only be 0, 4, 8, 12. + // For TEW = 64, mtd can only be 0, 2, 4, 6, 8, 10, 12, 14. + // Only `sf_mm_f_f_w1` and `sf_mm_f_f_w2` might have TEW = 64. + if ((BuiltinID == RISCVVector::BI__builtin_rvv_sf_mm_f_f_w1 && + EltSize == 64) || + (BuiltinID == RISCVVector::BI__builtin_rvv_sf_mm_f_f_w2 && + EltSize == 32)) + return SemaRef.BuiltinConstantArgRange(TheCall, 0, 0, 15) || + SemaRef.BuiltinConstantArgMultiple(TheCall, 0, 2); + return SemaRef.BuiltinConstantArgRange(TheCall, 0, 0, 15) || + SemaRef.BuiltinConstantArgMultiple(TheCall, 0, 4); + } + case RISCVVector::BI__builtin_rvv_sf_vtzero_t: { + llvm::APSInt Log2SEWResult; + llvm::APSInt TWidenResult; + if (SemaRef.BuiltinConstantArg(TheCall, 3, Log2SEWResult) || + SemaRef.BuiltinConstantArg(TheCall, 4, TWidenResult)) + return true; + + int Log2SEW = Log2SEWResult.getSExtValue(); + int TWiden = TWidenResult.getSExtValue(); + + // 3 <= LogSEW <= 6 + if (SemaRef.BuiltinConstantArgRange(TheCall, 3, 3, 6)) + return true; + + // TWiden + if (TWiden != 1 && TWiden != 2 && TWiden != 4) + return Diag(TheCall->getBeginLoc(), + diag::err_riscv_builtin_invalid_twiden); + + int TEW = (1 << Log2SEW) * TWiden; + + // For TEW = 8, mtd can be 0~15. + // For TEW = 16 or 64, mtd can only be 0, 2, 4, 6, 8, 10, 12, 14. + // For TEW = 32, mtd can only be 0, 4, 8, 12. + if (SemaRef.BuiltinConstantArgRange(TheCall, 0, 0, 15)) + return true; + if (TEW == 16 || TEW == 64) + return SemaRef.BuiltinConstantArgMultiple(TheCall, 0, 2); + return SemaRef.BuiltinConstantArgMultiple(TheCall, 0, 4); + } case RISCVVector::BI__builtin_rvv_vget_v: { ASTContext::BuiltinVectorTypeInfo ResVecInfo = Context.getBuiltinVectorTypeInfo(cast<BuiltinType>( diff --git a/clang/lib/StaticAnalyzer/Checkers/WebKit/RawPtrRefLambdaCapturesChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/WebKit/RawPtrRefLambdaCapturesChecker.cpp index 033eb8c..f60d193 100644 --- a/clang/lib/StaticAnalyzer/Checkers/WebKit/RawPtrRefLambdaCapturesChecker.cpp +++ b/clang/lib/StaticAnalyzer/Checkers/WebKit/RawPtrRefLambdaCapturesChecker.cpp @@ -50,7 +50,9 @@ public: llvm::DenseSet<const DeclRefExpr *> DeclRefExprsToIgnore; llvm::DenseSet<const LambdaExpr *> LambdasToIgnore; llvm::DenseSet<const ValueDecl *> ProtectedThisDecls; + llvm::DenseSet<const CallExpr *> CallToIgnore; llvm::DenseSet<const CXXConstructExpr *> ConstructToIgnore; + llvm::DenseMap<const VarDecl *, const LambdaExpr *> LambdaOwnerMap; QualType ClsType; @@ -101,10 +103,60 @@ public: auto *Init = VD->getInit(); if (!Init) return true; - auto *L = dyn_cast_or_null<LambdaExpr>(Init->IgnoreParenCasts()); - if (!L) + if (auto *L = dyn_cast_or_null<LambdaExpr>(Init->IgnoreParenCasts())) { + LambdasToIgnore.insert(L); // Evaluate lambdas in VisitDeclRefExpr. + return true; + } + if (!VD->hasLocalStorage()) return true; - LambdasToIgnore.insert(L); // Evaluate lambdas in VisitDeclRefExpr. + if (auto *E = dyn_cast<ExprWithCleanups>(Init)) + Init = E->getSubExpr(); + if (auto *E = dyn_cast<CXXBindTemporaryExpr>(Init)) + Init = E->getSubExpr(); + if (auto *CE = dyn_cast<CallExpr>(Init)) { + if (auto *Callee = CE->getDirectCallee()) { + auto FnName = safeGetName(Callee); + unsigned ArgCnt = CE->getNumArgs(); + if (FnName == "makeScopeExit" && ArgCnt == 1) { + auto *Arg = CE->getArg(0); + if (auto *E = dyn_cast<MaterializeTemporaryExpr>(Arg)) + Arg = E->getSubExpr(); + if (auto *L = dyn_cast<LambdaExpr>(Arg)) { + LambdaOwnerMap.insert(std::make_pair(VD, L)); + CallToIgnore.insert(CE); + LambdasToIgnore.insert(L); + } + } else if (FnName == "makeVisitor") { + for (unsigned ArgIndex = 0; ArgIndex < ArgCnt; ++ArgIndex) { + auto *Arg = CE->getArg(ArgIndex); + if (auto *E = dyn_cast<MaterializeTemporaryExpr>(Arg)) + Arg = E->getSubExpr(); + if (auto *L = dyn_cast<LambdaExpr>(Arg)) { + LambdaOwnerMap.insert(std::make_pair(VD, L)); + CallToIgnore.insert(CE); + LambdasToIgnore.insert(L); + } + } + } + } + } else if (auto *CE = dyn_cast<CXXConstructExpr>(Init)) { + if (auto *Ctor = CE->getConstructor()) { + if (auto *Cls = Ctor->getParent()) { + auto FnName = safeGetName(Cls); + unsigned ArgCnt = CE->getNumArgs(); + if (FnName == "ScopeExit" && ArgCnt == 1) { + auto *Arg = CE->getArg(0); + if (auto *E = dyn_cast<MaterializeTemporaryExpr>(Arg)) + Arg = E->getSubExpr(); + if (auto *L = dyn_cast<LambdaExpr>(Arg)) { + LambdaOwnerMap.insert(std::make_pair(VD, L)); + ConstructToIgnore.insert(CE); + LambdasToIgnore.insert(L); + } + } + } + } + } return true; } @@ -114,6 +166,12 @@ public: auto *VD = dyn_cast_or_null<VarDecl>(DRE->getDecl()); if (!VD) return true; + if (auto It = LambdaOwnerMap.find(VD); It != LambdaOwnerMap.end()) { + auto *L = It->second; + Checker->visitLambdaExpr(L, shouldCheckThis() && !hasProtectedThis(L), + ClsType); + return true; + } auto *Init = VD->getInit(); if (!Init) return true; @@ -167,10 +225,14 @@ public: } bool VisitCallExpr(CallExpr *CE) override { + if (CallToIgnore.contains(CE)) + return true; checkCalleeLambda(CE); - if (auto *Callee = CE->getDirectCallee()) + if (auto *Callee = CE->getDirectCallee()) { + if (isVisitFunction(CE, Callee)) + return true; checkParameters(CE, Callee); - else if (auto *CalleeE = CE->getCallee()) { + } else if (auto *CalleeE = CE->getCallee()) { if (auto *DRE = dyn_cast<DeclRefExpr>(CalleeE->IgnoreParenCasts())) { if (auto *Callee = dyn_cast_or_null<FunctionDecl>(DRE->getDecl())) checkParameters(CE, Callee); @@ -179,6 +241,34 @@ public: return true; } + bool isVisitFunction(CallExpr *CallExpr, FunctionDecl *FnDecl) { + bool IsVisitFn = safeGetName(FnDecl) == "visit"; + if (!IsVisitFn) + return false; + bool ArgCnt = CallExpr->getNumArgs(); + if (!ArgCnt) + return false; + auto *Ns = FnDecl->getParent(); + if (!Ns) + return false; + auto NsName = safeGetName(Ns); + if (NsName != "WTF" && NsName != "std") + return false; + auto *Arg = CallExpr->getArg(0); + if (!Arg) + return false; + auto *DRE = dyn_cast<DeclRefExpr>(Arg->IgnoreParenCasts()); + if (!DRE) + return false; + auto *VD = dyn_cast<VarDecl>(DRE->getDecl()); + if (!VD) + return false; + if (!LambdaOwnerMap.contains(VD)) + return false; + DeclRefExprsToIgnore.insert(DRE); + return true; + } + void checkParameters(CallExpr *CE, FunctionDecl *Callee) { unsigned ArgIndex = isa<CXXOperatorCallExpr>(CE); bool TreatAllArgsAsNoEscape = shouldTreatAllArgAsNoEscape(Callee); @@ -280,7 +370,7 @@ public: LambdasToIgnore.insert(L); } - bool hasProtectedThis(LambdaExpr *L) { + bool hasProtectedThis(const LambdaExpr *L) { for (const LambdaCapture &OtherCapture : L->captures()) { if (!OtherCapture.capturesVariable()) continue; @@ -378,7 +468,8 @@ public: visitor.TraverseDecl(const_cast<TranslationUnitDecl *>(TUD)); } - void visitLambdaExpr(LambdaExpr *L, bool shouldCheckThis, const QualType T, + void visitLambdaExpr(const LambdaExpr *L, bool shouldCheckThis, + const QualType T, bool ignoreParamVarDecl = false) const { if (TFA.isTrivial(L->getBody())) return; @@ -410,7 +501,7 @@ public: } void reportBug(const LambdaCapture &Capture, ValueDecl *CapturedVar, - const QualType T, LambdaExpr *L) const { + const QualType T, const LambdaExpr *L) const { assert(CapturedVar); auto Location = Capture.getLocation(); diff --git a/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp b/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp index 82b560b..e0deec1 100644 --- a/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp +++ b/clang/lib/StaticAnalyzer/Frontend/AnalysisConsumer.cpp @@ -128,7 +128,6 @@ public: std::unique_ptr<llvm::Timer> SyntaxCheckTimer; std::unique_ptr<llvm::Timer> ExprEngineTimer; std::unique_ptr<llvm::Timer> BugReporterTimer; - bool ShouldClearTimersToPreventDisplayingThem; /// The information about analyzed functions shared throughout the /// translation unit. @@ -149,7 +148,10 @@ public: if (Opts.AnalyzerDisplayProgress || Opts.PrintStats || Opts.ShouldSerializeStats || !Opts.DumpEntryPointStatsToCSV.empty()) { AnalyzerTimers = std::make_unique<llvm::TimerGroup>( - "analyzer", "Analyzer timers"); + "analyzer", "Analyzer timers", + /*PrintOnExit=*/ + (Opts.AnalyzerDisplayProgress || Opts.PrintStats || + Opts.ShouldSerializeStats)); SyntaxCheckTimer = std::make_unique<llvm::Timer>( "syntaxchecks", "Syntax-based analysis time", *AnalyzerTimers); ExprEngineTimer = std::make_unique<llvm::Timer>( @@ -159,12 +161,6 @@ public: *AnalyzerTimers); } - // Avoid displaying the timers created above in case we only want to record - // per-entry-point stats. - ShouldClearTimersToPreventDisplayingThem = !Opts.AnalyzerDisplayProgress && - !Opts.PrintStats && - !Opts.ShouldSerializeStats; - if (Opts.PrintStats || Opts.ShouldSerializeStats) { llvm::EnableStatistics(/* DoPrintOnExit= */ false); } @@ -287,9 +283,6 @@ public: checkerMgr->runCheckersOnASTDecl(D, *Mgr, *RecVisitorBR); if (SyntaxCheckTimer) SyntaxCheckTimer->stopTimer(); - if (AnalyzerTimers && ShouldClearTimersToPreventDisplayingThem) { - AnalyzerTimers->clear(); - } } return true; } @@ -583,9 +576,6 @@ void AnalysisConsumer::runAnalysisOnTranslationUnit(ASTContext &C) { checkerMgr->runCheckersOnASTDecl(TU, *Mgr, BR); if (SyntaxCheckTimer) SyntaxCheckTimer->stopTimer(); - if (AnalyzerTimers && ShouldClearTimersToPreventDisplayingThem) { - AnalyzerTimers->clear(); - } // Run the AST-only checks using the order in which functions are defined. // If inlining is not turned on, use the simplest function order for path @@ -765,9 +755,6 @@ void AnalysisConsumer::HandleCode(Decl *D, AnalysisMode Mode, FunctionSummaries.findOrInsertSummary(D)->second.SyntaxRunningTime = std::lround(CheckerDuration.getWallTime() * 1000); DisplayTime(CheckerDuration); - if (AnalyzerTimers && ShouldClearTimersToPreventDisplayingThem) { - AnalyzerTimers->clear(); - } } } @@ -830,9 +817,6 @@ void AnalysisConsumer::RunPathSensitiveChecks(Decl *D, PathRunningTime.set(static_cast<unsigned>( std::lround(ExprEngineDuration.getWallTime() * 1000))); DisplayTime(ExprEngineDuration); - if (AnalyzerTimers && ShouldClearTimersToPreventDisplayingThem) { - AnalyzerTimers->clear(); - } } if (!Mgr->options.DumpExplodedGraphTo.empty()) @@ -843,9 +827,6 @@ void AnalysisConsumer::RunPathSensitiveChecks(Decl *D, Eng.ViewGraph(Mgr->options.TrimGraph); flushReports(BugReporterTimer.get(), Eng.getBugReporter()); - if (AnalyzerTimers && ShouldClearTimersToPreventDisplayingThem) { - AnalyzerTimers->clear(); - } } //===----------------------------------------------------------------------===// diff --git a/clang/lib/Support/RISCVVIntrinsicUtils.cpp b/clang/lib/Support/RISCVVIntrinsicUtils.cpp index dad3d0da..12e209a 100644 --- a/clang/lib/Support/RISCVVIntrinsicUtils.cpp +++ b/clang/lib/Support/RISCVVIntrinsicUtils.cpp @@ -980,11 +980,12 @@ RVVIntrinsic::RVVIntrinsic( bool HasMaskedOffOperand, bool HasVL, PolicyScheme Scheme, bool SupportOverloading, bool HasBuiltinAlias, StringRef ManualCodegen, const RVVTypes &OutInTypes, const std::vector<int64_t> &NewIntrinsicTypes, - unsigned NF, Policy NewPolicyAttrs, bool HasFRMRoundModeOp) + unsigned NF, Policy NewPolicyAttrs, bool HasFRMRoundModeOp, unsigned TWiden) : IRName(IRName), IsMasked(IsMasked), HasMaskedOffOperand(HasMaskedOffOperand), HasVL(HasVL), Scheme(Scheme), SupportOverloading(SupportOverloading), HasBuiltinAlias(HasBuiltinAlias), - ManualCodegen(ManualCodegen.str()), NF(NF), PolicyAttrs(NewPolicyAttrs) { + ManualCodegen(ManualCodegen.str()), NF(NF), PolicyAttrs(NewPolicyAttrs), + TWiden(TWiden) { // Init BuiltinName, Name and OverloadedName BuiltinName = NewName.str(); diff --git a/clang/test/AST/ByteCode/arrays.cpp b/clang/test/AST/ByteCode/arrays.cpp index 22a4b41..eaf9559 100644 --- a/clang/test/AST/ByteCode/arrays.cpp +++ b/clang/test/AST/ByteCode/arrays.cpp @@ -820,3 +820,14 @@ namespace FAM { return 1; } } + +namespace MultiDimConstructExpr { + struct a { + a *p = this; + }; + struct b { + a m[3][3]; + }; + constexpr b d; + static_assert(d.m[2][1].p == &d.m[2][1]); +} diff --git a/clang/test/AST/ByteCode/placement-new.cpp b/clang/test/AST/ByteCode/placement-new.cpp index b587cd6..1c015da 100644 --- a/clang/test/AST/ByteCode/placement-new.cpp +++ b/clang/test/AST/ByteCode/placement-new.cpp @@ -494,3 +494,32 @@ constexpr int modify_const_variable() { } static_assert(modify_const_variable()); // both-error {{not an integral constant expression}} \ // both-note {{in call to}} + +constexpr int nullDest() { + new (nullptr) int{12}; // both-note {{construction of dereferenced null pointer}} + return 0; +} +static_assert(nullDest() == 0); // both-error {{not an integral constant expression}} \ + // both-note {{in call to}} + +constexpr int nullArrayDest() { + new (nullptr) int{12}; // both-note {{construction of dereferenced null pointer}} + return 0; +} +static_assert(nullArrayDest() == 0); // both-error {{not an integral constant expression}} \ + // both-note {{in call to}} + +constexpr int intDest() { + new ((void*)2) int{3}; // both-note {{cast that performs the conversions of a reinterpret_cast}} + return 0; +} +static_assert(intDest() == 0); // both-error {{not an integral constant expression}} \ + // both-note {{in call to}} + +constexpr int intDestArray() { + new ((void*)2) int[4]; // both-note {{cast that performs the conversions of a reinterpret_cast}} + return 0; +} +static_assert(intDestArray() == 0); // both-error {{not an integral constant expression}} \ + // both-note {{in call to}} + diff --git a/clang/test/AST/HLSL/matrix-constructors.hlsl b/clang/test/AST/HLSL/matrix-constructors.hlsl new file mode 100644 index 0000000..0a2f03c --- /dev/null +++ b/clang/test/AST/HLSL/matrix-constructors.hlsl @@ -0,0 +1,393 @@ +// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.0-compute -x hlsl -ast-dump -o - %s | FileCheck %s + +typedef float float2x1 __attribute__((matrix_type(2,1))); +typedef float float2x3 __attribute__((matrix_type(2,3))); +typedef float float2x2 __attribute__((matrix_type(2,2))); +typedef float float4x4 __attribute__((matrix_type(4,4))); +typedef float float2 __attribute__((ext_vector_type(2))); +typedef float float4 __attribute__((ext_vector_type(4))); + +[numthreads(1,1,1)] +void ok() { + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} A 'float2x3':'matrix<float, 2, 3>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' functional cast to float2x3 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 5 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 6 + float2x3 A = float2x3(1,2,3,4,5,6); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} B 'float2x1':'matrix<float, 2, 1>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' functional cast to float2x1 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 1.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 2.000000e+00 + float2x1 B = float2x1(1.0,2.0); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} C 'float2x1':'matrix<float, 2, 1>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' functional cast to float2x1 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' +// CHECK-NEXT: UnaryOperator 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' prefix '-' +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 1.000000e+00 +// CHECK-NEXT: UnaryOperator 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' prefix '-' +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 2.000000e+00 + float2x1 C = float2x1(-1.0f,-2.0f); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} D 'float2x3':'matrix<float, 2, 3>' cinit +// CHECK-NEXT: ExprWithCleanups 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' functional cast to float2x3 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 5 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 6 + float2x3 D = float2x3(float2(1,2), 3, 4, 5, 6); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} E 'float2x3':'matrix<float, 2, 3>' cinit +// CHECK-NEXT: ExprWithCleanups 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' functional cast to float2x3 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' functional cast to float2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 5 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 6 + float2x3 E = float2x3(float2(1,2), float2(3,4), 5, 6); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} F 'float2x3':'matrix<float, 2, 3>' cinit +// CHECK-NEXT: ExprWithCleanups 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' functional cast to float2x3 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' functional cast to float4 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' functional cast to float4 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' functional cast to float4 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' xvalue +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' functional cast to float4 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4':'vector<float, 4>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 5 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 6 + float2x3 F = float2x3(float4(1,2,3,4), 5, 6); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} G 'float2x3':'matrix<float, 2, 3>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' functional cast to float2x3 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x3':'matrix<float, 2, 3>' +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' matrixcomponent +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' matrixcomponent +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' matrixcomponent +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' matrixcomponent +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 5 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 6 +float2x3 G = float2x3(float2x2(1,2,3,4), 5, 6); + + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} H 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'Vec2' 'float2':'vector<float, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'Vec2' 'float2':'vector<float, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 + float2 Vec2 = float2(1.0, 2.0); + float2x2 H = float2x2(Vec2,3,4); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} I 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' lvalue Var 0x{{[0-9a-fA-F]+}} 'i' 'int' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' lvalue Var 0x{{[0-9a-fA-F]+}} 'j' 'int' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' lvalue Var 0x{{[0-9a-fA-F]+}} 'k' 'int' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' lvalue Var 0x{{[0-9a-fA-F]+}} 'l' 'int' + int i = 1, j = 2, k = 3, l = 4; + float2x2 I = float2x2(i,j,k,l); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} J 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'struct S' lvalue Var 0x{{[0-9a-fA-F]+}} 's' 'struct S' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'struct S' lvalue Var 0x{{[0-9a-fA-F]+}} 's' 'struct S' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' lvalue .a 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'struct S' lvalue Var 0x{{[0-9a-fA-F]+}} 's' 'struct S' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' lvalue .a 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'struct S' lvalue Var 0x{{[0-9a-fA-F]+}} 's' 'struct S' + struct S { float2 f; float a;} s; + float2x2 J = float2x2(s.f, s.a, s.a); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} L 'second_level_of_typedefs':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 1.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 2.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 3.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 4.000000e+00 + typedef float2x2 second_level_of_typedefs; + second_level_of_typedefs L = float2x2(1.0f, 2.0f, 3.0f, 4.0f); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} M 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'second_level_of_typedefs':'matrix<float, 2, 2>' functional cast to second_level_of_typedefs <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'second_level_of_typedefs':'matrix<float, 2, 2>' +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 1.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 2.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 3.000000e+00 +// CHECK-NEXT: FloatingLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' 4.000000e+00 + float2x2 M = second_level_of_typedefs(1.0f, 2.0f, 3.0f, 4.0f); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} N 'float4x4':'matrix<float, 4, 4>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4x4':'matrix<float, 4, 4>' functional cast to float4x4 <HLSLElementwiseCast> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'sF' lvalue Var 0x{{[0-9a-fA-F]+}} 'f' 'sF' +struct sF { + float f[16]; +}; + +sF f; +float4x4 N = float4x4(f); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} GettingStrange 'float2x1':'matrix<float, 2, 1>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' functional cast to float2x1 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x1':'matrix<float, 2, 1>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S2' lvalue Var 0x{{[0-9a-fA-F]+}} 's2' 'S2' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S2' lvalue Var 0x{{[0-9a-fA-F]+}} 's2' 'S2' +struct S2 { float f; }; +S2 s2; +float2x1 GettingStrange = float2x1(s2, s2); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} GettingStrange2 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <NoOp> +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S3' lvalue Var 0x{{[0-9a-fA-F]+}} 's3' 'S3' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S3' lvalue Var 0x{{[0-9a-fA-F]+}} 's3' 'S3' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S3' lvalue Var 0x{{[0-9a-fA-F]+}} 's3' 'S3' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue vectorcomponent +// CHECK-NEXT: MemberExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2':'vector<float, 2>' lvalue .f 0x{{[0-9a-fA-F]+}} +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S3' lvalue Var 0x{{[0-9a-fA-F]+}} 's3' 'S3' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +struct S3 { float2 f;}; +S3 s3; +float2x2 GettingStrange2 = float2x2(s3, s3); + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} GettingStrange3 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: CXXFunctionalCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' functional cast to float2x2 <HLSLElementwiseCast> +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'S4' lvalue Var 0x{{[0-9a-fA-F]+}} 's4' 'S4' +struct S4 { float4 f;}; +S4 s4; +float2x2 GettingStrange3 = float2x2(s4); + +} + diff --git a/clang/test/AST/HLSL/matrix-general-initializer.hlsl b/clang/test/AST/HLSL/matrix-general-initializer.hlsl new file mode 100644 index 0000000..14c950a --- /dev/null +++ b/clang/test/AST/HLSL/matrix-general-initializer.hlsl @@ -0,0 +1,260 @@ +// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.0-compute -x hlsl -ast-dump -o - %s | FileCheck %s + +typedef float float4x2 __attribute__((matrix_type(4,2))); +typedef float float2x2 __attribute__((matrix_type(2,2))); +typedef int int4x4 __attribute__((matrix_type(4,4))); + + +[numthreads(1,1,1)] +void ok() { + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} used m 'float4x2':'matrix<float, 4, 2>' cinit +// CHECK-NEXT: ExprWithCleanups 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 3>' xxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' x +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 2>' xx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 4 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +float4x2 m = {1.xxx, 2.xx, 3.x, 4.xx}; + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} s 'S' cinit +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'S' +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 2 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float4x2':'matrix<float, 4, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm' 'float4x2':'matrix<float, 4, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 3 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +struct S { float2x2 x; float2x2 y;}; +S s = {m}; + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} used m2 'float2x2':'matrix<float, 2, 2>' cinit +// CHECK-NEXT: ExprWithCleanups 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xxxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xxxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xxxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 2 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'float' <IntegralToFloating> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' <LValueToRValue> +// CHECK-NEXT: ArraySubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int' xvalue vectorcomponent +// CHECK-NEXT: MaterializeTemporaryExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xvalue +// CHECK-NEXT: ExtVectorElementExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'vector<int, 4>' xxxx +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'vector<int, 1>' <VectorSplat> +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <<invalid sloc>> '__size_t':'unsigned long' 3 +float2x2 m2 = {0.xxxx}; + +// CHECK: VarDecl 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> col:{{[0-9]+}} m3 'int4x4':'matrix<int, 4, 4>' cinit +// CHECK-NEXT: InitListExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}, col:{{[0-9]+}}> 'int4x4':'matrix<int, 4, 4>' +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 0 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' <FloatingToIntegral> +// CHECK-NEXT: ImplicitCastExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' <LValueToRValue> +// CHECK-NEXT: MatrixSubscriptExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float' lvalue matrixcomponent +// CHECK-NEXT: DeclRefExpr 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'float2x2':'matrix<float, 2, 2>' lvalue Var 0x{{[0-9a-fA-F]+}} 'm2' 'float2x2':'matrix<float, 2, 2>' +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +// CHECK-NEXT: IntegerLiteral 0x{{[0-9a-fA-F]+}} <col:{{[0-9]+}}> 'int' 1 +int4x4 m3 = {m2, m2, m2, m2}; + +}
\ No newline at end of file diff --git a/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures.cpp b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures.cpp index a4ad741..fd1eecd 100644 --- a/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures.cpp +++ b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures.cpp @@ -107,6 +107,79 @@ private: ValueType* m_table { nullptr }; }; +class ScopeExit final { +public: + template<typename ExitFunctionParameter> + explicit ScopeExit(ExitFunctionParameter&& exitFunction) + : m_exitFunction(std::move(exitFunction)) { + } + + ScopeExit(ScopeExit&& other) + : m_exitFunction(std::move(other.m_exitFunction)) + , m_execute(std::move(other.m_execute)) { + } + + ~ScopeExit() { + if (m_execute) + m_exitFunction(); + } + + WTF::Function<void()> take() { + m_execute = false; + return std::move(m_exitFunction); + } + + void release() { m_execute = false; } + + ScopeExit(const ScopeExit&) = delete; + ScopeExit& operator=(const ScopeExit&) = delete; + ScopeExit& operator=(ScopeExit&&) = delete; + +private: + WTF::Function<void()> m_exitFunction; + bool m_execute { true }; +}; + +template<typename ExitFunction> ScopeExit makeScopeExit(ExitFunction&&); +template<typename ExitFunction> +ScopeExit makeScopeExit(ExitFunction&& exitFunction) +{ + return ScopeExit(std::move(exitFunction)); +} + +// Visitor adapted from http://stackoverflow.com/questions/25338795/is-there-a-name-for-this-tuple-creation-idiom + +template<class A, class... B> struct Visitor : Visitor<A>, Visitor<B...> { + Visitor(A a, B... b) + : Visitor<A>(a) + , Visitor<B...>(b...) + { + } + + using Visitor<A>::operator (); + using Visitor<B...>::operator (); +}; + +template<class A> struct Visitor<A> : A { + Visitor(A a) + : A(a) + { + } + + using A::operator(); +}; + +template<class... F> Visitor<F...> makeVisitor(F... f) +{ + return Visitor<F...>(f...); +} + +void opaqueFunction(); +template <typename Visitor, typename... Variants> void visit(Visitor&& v, Variants&&... values) +{ + opaqueFunction(); +} + } // namespace WTF struct A { @@ -501,3 +574,81 @@ void RefCountedObj::call() const }; callLambda(lambda); } + +void scope_exit(RefCountable* obj) { + auto scope = WTF::makeScopeExit([&] { + obj->method(); + }); + someFunction(); + WTF::ScopeExit scope2([&] { + obj->method(); + }); + someFunction(); +} + +void doWhateverWith(WTF::ScopeExit& obj); + +void scope_exit_with_side_effect(RefCountable* obj) { + auto scope = WTF::makeScopeExit([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); + doWhateverWith(scope); +} + +void scope_exit_static(RefCountable* obj) { + static auto scope = WTF::makeScopeExit([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); +} + +WTF::Function<void()> scope_exit_take_lambda(RefCountable* obj) { + auto scope = WTF::makeScopeExit([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); + return scope.take(); +} + +// FIXME: Ideally, we treat release() as a trivial function. +void scope_exit_release(RefCountable* obj) { + auto scope = WTF::makeScopeExit([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); + scope.release(); +} + +void make_visitor(RefCountable* obj) { + auto visitor = WTF::makeVisitor([&] { + obj->method(); + }); +} + +void use_visitor(RefCountable* obj) { + auto visitor = WTF::makeVisitor([&] { + obj->method(); + }); + WTF::visit(visitor, obj); +} + +template <typename Visitor, typename ObjectType> +void bad_visit(Visitor&, ObjectType*) { + someFunction(); +} + +void static_visitor(RefCountable* obj) { + static auto visitor = WTF::makeVisitor([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); +} + +void bad_use_visitor(RefCountable* obj) { + auto visitor = WTF::makeVisitor([&] { + obj->method(); + // expected-warning@-1{{Implicitly captured raw-pointer 'obj' to uncounted type is unsafe [webkit.UncountedLambdaCapturesChecker]}} + }); + bad_visit(visitor, obj); +} diff --git a/clang/test/CIR/CodeGen/complex.cpp b/clang/test/CIR/CodeGen/complex.cpp index 4e89af4..3fb78dc 100644 --- a/clang/test/CIR/CodeGen/complex.cpp +++ b/clang/test/CIR/CodeGen/complex.cpp @@ -1468,3 +1468,30 @@ void calling_function_with_default_arg() { // OGCG: store float 0x40019999A0000000, ptr %[[DEFAULT_ARG_IMAG_PTR]], align 4 // OGCG: %[[TMP_DEFAULT_ARG:.*]] = load <2 x float>, ptr %[[DEFAULT_ARG_ADDR]], align 4 // OGCG: call void @_Z33function_with_complex_default_argCf(<2 x float> {{.*}} %[[TMP_DEFAULT_ARG]]) + +void calling_function_that_return_complex() { + float _Complex a = complex_type_return_type(); +} + +// CIR: %[[A_ADDR:.*]] = cir.alloca !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>>, ["a", init] +// CIR: %[[RESULT:.*]] = cir.call @_Z24complex_type_return_typev() : () -> !cir.complex<!cir.float> +// CIR: cir.store{{.*}} %[[RESULT]], %[[A_ADDR]] : !cir.complex<!cir.float>, !cir.ptr<!cir.complex<!cir.float>> + +// TODO(CIR): the difference between the CIR LLVM and OGCG is because the lack of calling convention lowering, + +// LLVM: %[[A_ADDR:.*]] = alloca { float, float }, i64 1, align 4 +// LLVM: %[[RESULT:.*]] = call { float, float } @_Z24complex_type_return_typev() +// LLVM: store { float, float } %[[RESULT]], ptr %[[A_ADDR]], align 4 + +// OGCG: %[[A_ADDR:.*]] = alloca { float, float }, align 4 +// OGCG: %[[RESULT_ADDR:.*]] = alloca { float, float }, align 4 +// OGCG: %[[RESULT:.*]] = call noundef <2 x float> @_Z24complex_type_return_typev() +// OGCG: store <2 x float> %[[RESULT]], ptr %[[RESULT_ADDR]], align 4 +// OGCG: %[[RESULT_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[RESULT_ADDR]], i32 0, i32 0 +// OGCG: %[[RESULT_REAL:.*]] = load float, ptr %[[RESULT_REAL_PTR]], align 4 +// OGCG: %[[RESULT_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[RESULT_ADDR]], i32 0, i32 1 +// OGCG: %[[RESULT_IMAG:.*]] = load float, ptr %[[RESULT_IMAG_PTR]], align 4 +// OGCG: %[[A_REAL_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 0 +// OGCG: %[[A_IMAG_PTR:.*]] = getelementptr inbounds nuw { float, float }, ptr %[[A_ADDR]], i32 0, i32 1 +// OGCG: store float %[[RESULT_REAL]], ptr %[[A_REAL_PTR]], align 4 +// OGCG: store float %[[RESULT_IMAG]], ptr %[[A_IMAG_PTR]], align 4 diff --git a/clang/test/CIR/CodeGen/coro-task.cpp b/clang/test/CIR/CodeGen/coro-task.cpp index 1fc7d77..265325f 100644 --- a/clang/test/CIR/CodeGen/coro-task.cpp +++ b/clang/test/CIR/CodeGen/coro-task.cpp @@ -106,6 +106,9 @@ co_invoke_fn co_invoke; // CIR-NEXT: cir.global external @_ZN5folly4coro9co_invokeE = #cir.zero : !rec_folly3A3Acoro3A3Aco_invoke_fn // CIR: cir.func builtin private @__builtin_coro_id(!u32i, !cir.ptr<!void>, !cir.ptr<!void>, !cir.ptr<!void>) -> !u32i +// CIR: cir.func builtin private @__builtin_coro_alloc(!u32i) -> !cir.bool +// CIR: cir.func builtin private @__builtin_coro_size() -> !u64i +// CIR: cir.func builtin private @__builtin_coro_begin(!u32i, !cir.ptr<!void>) -> !cir.ptr<!void> using VoidTask = folly::coro::Task<void>; @@ -114,10 +117,24 @@ VoidTask silly_task() { } // CIR: cir.func coroutine dso_local @_Z10silly_taskv() -> ![[VoidTask]] -// CHECK: %[[#VoidTaskAddr:]] = cir.alloca ![[VoidTask]], {{.*}}, ["__retval"] +// CIR: %[[VoidTaskAddr:.*]] = cir.alloca ![[VoidTask]], {{.*}}, ["__retval"] +// CIR: %[[SavedFrameAddr:.*]] = cir.alloca !cir.ptr<!void>, !cir.ptr<!cir.ptr<!void>>, ["__coro_frame_addr"] // Get coroutine id with __builtin_coro_id. // CIR: %[[NullPtr:.*]] = cir.const #cir.ptr<null> : !cir.ptr<!void> // CIR: %[[Align:.*]] = cir.const #cir.int<16> : !u32i // CIR: %[[CoroId:.*]] = cir.call @__builtin_coro_id(%[[Align]], %[[NullPtr]], %[[NullPtr]], %[[NullPtr]]) + +// Perform allocation calling operator 'new' depending on __builtin_coro_alloc and +// call __builtin_coro_begin for the final coroutine frame address. + +// CIR: %[[ShouldAlloc:.*]] = cir.call @__builtin_coro_alloc(%[[CoroId]]) : (!u32i) -> !cir.bool +// CIR: cir.store{{.*}} %[[NullPtr]], %[[SavedFrameAddr]] : !cir.ptr<!void>, !cir.ptr<!cir.ptr<!void>> +// CIR: cir.if %[[ShouldAlloc]] { +// CIR: %[[CoroSize:.*]] = cir.call @__builtin_coro_size() : () -> !u64i +// CIR: %[[AllocAddr:.*]] = cir.call @_Znwm(%[[CoroSize]]) : (!u64i) -> !cir.ptr<!void> +// CIR: cir.store{{.*}} %[[AllocAddr]], %[[SavedFrameAddr]] : !cir.ptr<!void>, !cir.ptr<!cir.ptr<!void>> +// CIR: } +// CIR: %[[Load0:.*]] = cir.load{{.*}} %[[SavedFrameAddr]] : !cir.ptr<!cir.ptr<!void>>, !cir.ptr<!void> +// CIR: %[[CoroFrameAddr:.*]] = cir.call @__builtin_coro_begin(%[[CoroId]], %[[Load0]]) diff --git a/clang/test/CIR/CodeGen/delete.cpp b/clang/test/CIR/CodeGen/delete.cpp index 69640aa..d8ac436 100644 --- a/clang/test/CIR/CodeGen/delete.cpp +++ b/clang/test/CIR/CodeGen/delete.cpp @@ -86,3 +86,42 @@ Container::~Container() { delete contents; } // These functions are declared/defined below the calls in OGCG. // OGCG: define linkonce_odr void @_ZN8ContentsD2Ev // OGCG: declare void @_ZdlPvm(ptr noundef, i64 noundef) + +struct StructWithVirtualDestructor { + virtual ~StructWithVirtualDestructor(); +}; + +void destroy(StructWithVirtualDestructor *x) { + delete x; +} + +// CIR: cir.func {{.*}} @_Z7destroyP27StructWithVirtualDestructor(%[[X_ARG:.*]]: !cir.ptr<!rec_StructWithVirtualDestructor> {{.*}}) +// CIR: %[[X_ADDR:.*]] = cir.alloca !cir.ptr<!rec_StructWithVirtualDestructor> +// CIR: cir.store %[[X_ARG]], %[[X_ADDR]] +// CIR: %[[X:.*]] = cir.load{{.*}} %[[X_ADDR]] +// CIR: %[[VTABLE_PTR:.*]] = cir.vtable.get_vptr %[[X]] : !cir.ptr<!rec_StructWithVirtualDestructor> -> !cir.ptr<!cir.vptr> +// CIR: %[[VTABLE:.*]] = cir.load{{.*}} %[[VTABLE_PTR]] : !cir.ptr<!cir.vptr>, !cir.vptr +// CIR: %[[DTOR_FN_ADDR_PTR:.*]] = cir.vtable.get_virtual_fn_addr %[[VTABLE]][1] +// CIR: %[[DTOR_FN_ADDR:.*]] = cir.load{{.*}} %[[DTOR_FN_ADDR_PTR]] +// CIR: cir.call %[[DTOR_FN_ADDR]](%[[X]]) + +// LLVM: define {{.*}} void @_Z7destroyP27StructWithVirtualDestructor(ptr %[[X_ARG:.*]]) +// LLVM: %[[X_ADDR:.*]] = alloca ptr +// LLVM: store ptr %[[X_ARG]], ptr %[[X_ADDR]] +// LLVM: %[[X:.*]] = load ptr, ptr %[[X_ADDR]] +// LLVM: %[[VTABLE:.*]] = load ptr, ptr %[[X]] +// LLVM: %[[DTOR_FN_ADDR_PTR:.*]] = getelementptr inbounds ptr, ptr %[[VTABLE]], i32 1 +// LLVM: %[[DTOR_FN_ADDR:.*]] = load ptr, ptr %[[DTOR_FN_ADDR_PTR]] +// LLVM: call void %[[DTOR_FN_ADDR]](ptr %[[X]]) + +// OGCG: define {{.*}} void @_Z7destroyP27StructWithVirtualDestructor(ptr {{.*}} %[[X_ARG:.*]]) +// OGCG: %[[X_ADDR:.*]] = alloca ptr +// OGCG: store ptr %[[X_ARG]], ptr %[[X_ADDR]] +// OGCG: %[[X:.*]] = load ptr, ptr %[[X_ADDR]] +// OGCG: %[[ISNULL:.*]] = icmp eq ptr %[[X]], null +// OGCG: br i1 %[[ISNULL]], label %{{.*}}, label %[[DELETE_NOTNULL:.*]] +// OGCG: [[DELETE_NOTNULL]]: +// OGCG: %[[VTABLE:.*]] = load ptr, ptr %[[X]] +// OGCG: %[[DTOR_FN_ADDR_PTR:.*]] = getelementptr inbounds ptr, ptr %[[VTABLE]], i64 1 +// OGCG: %[[DTOR_FN_ADDR:.*]] = load ptr, ptr %[[DTOR_FN_ADDR_PTR]] +// OGCG: call void %[[DTOR_FN_ADDR]](ptr {{.*}} %[[X]]) diff --git a/clang/test/CIR/CodeGen/global-init.cpp b/clang/test/CIR/CodeGen/global-init.cpp index a76094b..01e2868 100644 --- a/clang/test/CIR/CodeGen/global-init.cpp +++ b/clang/test/CIR/CodeGen/global-init.cpp @@ -141,6 +141,34 @@ float _Complex a = {num, num}; // OGCG: store float %[[REAL]], ptr @a, align 4 // OGCG: store float %[[IMAG]], ptr getelementptr inbounds nuw ({ float, float }, ptr @a, i32 0, i32 1), align 4 +float fp; +int i = (int)fp; + +// CIR-BEFORE-LPP: cir.global external @i = ctor : !s32i { +// CIR-BEFORE-LPP: %0 = cir.get_global @i : !cir.ptr<!s32i> +// CIR-BEFORE-LPP: %1 = cir.get_global @fp : !cir.ptr<!cir.float> +// CIR-BEFORE-LPP: %2 = cir.load{{.*}} %1 : !cir.ptr<!cir.float>, !cir.float +// CIR-BEFORE-LPP: %3 = cir.cast float_to_int %2 : !cir.float -> !s32i +// CIR-BEFORE-LPP: cir.store{{.*}} %3, %0 : !s32i, !cir.ptr<!s32i> +// CIR-BEFORE-LPP: } + +// CIR: cir.func internal private @__cxx_global_var_init.4() +// CIR: %[[I_ADDR:.*]] = cir.get_global @i : !cir.ptr<!s32i> +// CIR: %[[FP_ADDR:.*]] = cir.get_global @fp : !cir.ptr<!cir.float> +// CIR: %[[TMP_FP:.*]] = cir.load{{.*}} %[[FP_ADDR]] : !cir.ptr<!cir.float>, !cir.float +// CIR: %[[FP_I32:.*]] = cir.cast float_to_int %[[TMP_FP]] : !cir.float -> !s32i +// CIR: cir.store{{.*}} %[[FP_I32]], %[[I_ADDR]] : !s32i, !cir.ptr<!s32i> + +// LLVM: define internal void @__cxx_global_var_init.4() +// LLVM: %[[TMP_FP:.*]] = load float, ptr @fp, align 4 +// LLVM: %[[FP_I32:.*]] = fptosi float %[[TMP_FP]] to i32 +// LLVM: store i32 %[[FP_I32]], ptr @i, align 4 + +// OGCG: define internal void @__cxx_global_var_init.4() {{.*}} section ".text.startup" +// OGCG: %[[TMP_FP:.*]] = load float, ptr @fp, align 4 +// OGCG: %[[FP_I32:.*]] = fptosi float %[[TMP_FP]] to i32 +// OGCG: store i32 %[[FP_I32]], ptr @i, align 4 + // Common init function for all globals with default priority // CIR: cir.func private @_GLOBAL__sub_I_[[FILENAME:.*]]() { @@ -148,15 +176,18 @@ float _Complex a = {num, num}; // CIR: cir.call @__cxx_global_var_init.1() : () -> () // CIR: cir.call @__cxx_global_var_init.2() : () -> () // CIR: cir.call @__cxx_global_var_init.3() : () -> () +// CIR: cir.call @__cxx_global_var_init.4() : () -> () // LLVM: define void @_GLOBAL__sub_I_[[FILENAME]]() // LLVM: call void @__cxx_global_var_init() // LLVM: call void @__cxx_global_var_init.1() // LLVM: call void @__cxx_global_var_init.2() // LLVM: call void @__cxx_global_var_init.3() +// LLVM: call void @__cxx_global_var_init.4() // OGCG: define internal void @_GLOBAL__sub_I_[[FILENAME]]() {{.*}} section ".text.startup" { // OGCG: call void @__cxx_global_var_init() // OGCG: call void @__cxx_global_var_init.1() // OGCG: call void @__cxx_global_var_init.2() // OGCG: call void @__cxx_global_var_init.3() +// OGCG: call void @__cxx_global_var_init.4() diff --git a/clang/test/CIR/CodeGen/trivial-ctor-const-init.cpp b/clang/test/CIR/CodeGen/trivial-ctor-const-init.cpp new file mode 100644 index 0000000..7429549 --- /dev/null +++ b/clang/test/CIR/CodeGen/trivial-ctor-const-init.cpp @@ -0,0 +1,44 @@ +// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++11 -fclangir -emit-cir %s -o %t.cir +// RUN: FileCheck --input-file=%t.cir %s --check-prefix=CIR +// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++11 -fclangir -emit-llvm %s -o %t-cir.ll +// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefix=LLVM +// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++11 -emit-llvm %s -o %t.ll +// RUN: FileCheck --input-file=%t.ll %s --check-prefix=OGCG + +struct StructWithDefaultCtor { + int n; +}; + +StructWithDefaultCtor defCtor = StructWithDefaultCtor(); + +// CIR: cir.global {{.*}} @defCtor = #cir.zero : !rec_StructWithDefaultCtor +// LLVM: @defCtor = global %struct.StructWithDefaultCtor zeroinitializer +// OGCG: @defCtor = global %struct.StructWithDefaultCtor zeroinitializer + +struct StructWithCtorArg { + double value; + StructWithCtorArg(const double& x) : value(x) {} +}; + +StructWithCtorArg withArg = 0.0; + +// CIR: cir.global {{.*}} @withArg = #cir.zero : !rec_StructWithCtorArg +// LLVM: @withArg = global %struct.StructWithCtorArg zeroinitializer +// OGCG: @withArg = global %struct.StructWithCtorArg zeroinitializer + +// CIR: cir.func {{.*}} @__cxx_global_var_init() +// CIR: %[[TMP0:.*]] = cir.alloca !cir.double, !cir.ptr<!cir.double>, ["ref.tmp0"] +// CIR: %[[WITH_ARG:.*]] = cir.get_global @withArg : !cir.ptr<!rec_StructWithCtorArg> +// CIR: %[[ZERO:.*]] = cir.const #cir.fp<0.000000e+00> : !cir.double +// CIR: cir.store{{.*}} %[[ZERO]], %[[TMP0]] : !cir.double, !cir.ptr<!cir.double> +// CIR: cir.call @_ZN17StructWithCtorArgC1ERKd(%[[WITH_ARG]], %[[TMP0]]) : (!cir.ptr<!rec_StructWithCtorArg>, !cir.ptr<!cir.double>) -> () + +// LLVM: define {{.*}} void @__cxx_global_var_init() +// LLVM: %[[TMP0:.*]] = alloca double +// LLVM: store double 0.000000e+00, ptr %[[TMP0]] +// LLVM: call void @_ZN17StructWithCtorArgC1ERKd(ptr @withArg, ptr %[[TMP0]]) + +// OGCG: define {{.*}} void @__cxx_global_var_init() +// OGCG: %[[TMP0:.*]] = alloca double +// OGCG: store double 0.000000e+00, ptr %[[TMP0]] +// OGCG: call void @_ZN17StructWithCtorArgC1ERKd(ptr {{.*}} @withArg, ptr {{.*}} %[[TMP0]]) diff --git a/clang/test/CIR/CodeGenOpenACC/atomic-update.cpp b/clang/test/CIR/CodeGenOpenACC/atomic-update.cpp new file mode 100644 index 0000000..7ab6b62 --- /dev/null +++ b/clang/test/CIR/CodeGenOpenACC/atomic-update.cpp @@ -0,0 +1,151 @@ +// RUN: %clang_cc1 -fopenacc -triple x86_64-linux-gnu -Wno-openacc-self-if-potential-conflict -emit-cir -fclangir -triple x86_64-linux-pc %s -o - | FileCheck %s + +struct HasOps { + operator float(); + int thing(); +}; + +void use(int x, unsigned int y, float f, HasOps ops) { + // CHECK: cir.func{{.*}}(%[[X_ARG:.*]]: !s32i{{.*}}, %[[Y_ARG:.*]]: !u32i{{.*}}, %[[F_ARG:.*]]: !cir.float{{.*}}){{.*}}, %[[OPS_ARG:.*]]: !rec_HasOps{{.*}}) { + // CHECK-NEXT: %[[X_ALLOCA:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["x", init] + // CHECK-NEXT: %[[Y_ALLOCA:.*]] = cir.alloca !u32i, !cir.ptr<!u32i>, ["y", init] + // CHECK-NEXT: %[[F_ALLOCA:.*]] = cir.alloca !cir.float, !cir.ptr<!cir.float>, ["f", init] + // CHECK-NEXT: %[[OPS_ALLOCA:.*]] = cir.alloca !rec_HasOps, !cir.ptr<!rec_HasOps>, ["ops", init] + // CHECK-NEXT: cir.store %[[X_ARG]], %[[X_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // CHECK-NEXT: cir.store %[[Y_ARG]], %[[Y_ALLOCA]] : !u32i, !cir.ptr<!u32i> + // CHECK-NEXT: cir.store %[[F_ARG]], %[[F_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // CHECK-NEXT: cir.store %[[OPS_ARG]], %[[OPS_ALLOCA]] : !rec_HasOps, !cir.ptr<!rec_HasOps> + + // CHECK-NEXT: acc.atomic.update %[[X_ALLOCA]] : !cir.ptr<!s32i> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !s32i{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[TEMP_LOAD]]) nsw : !s32i, !s32i + // CHECK-NEXT: cir.store {{.*}}%[[INC]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !s32i + // CHECK-NEXT: } +#pragma acc atomic update + ++x; + + // CHECK-NEXT: acc.atomic.update %[[Y_ALLOCA]] : !cir.ptr<!u32i> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !u32i{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !u32i, !cir.ptr<!u32i>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !u32i, !cir.ptr<!u32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!u32i>, !u32i + // CHECK-NEXT: %[[INC:.*]] = cir.unary(inc, %[[TEMP_LOAD]]) : !u32i, !u32i + // CHECK-NEXT: cir.store {{.*}}%[[INC]], %[[TEMP_ALLOCA]] : !u32i, !cir.ptr<!u32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!u32i>, !u32i + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !u32i + // CHECK-NEXT: } +#pragma acc atomic update + y++; + + // CHECK-NEXT: acc.atomic.update %[[F_ALLOCA]] : !cir.ptr<!cir.float> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !cir.float{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !cir.float, !cir.ptr<!cir.float>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: %[[INC:.*]] = cir.unary(dec, %[[TEMP_LOAD]]) : !cir.float, !cir.float + // CHECK-NEXT: cir.store {{.*}}%[[INC]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !cir.float + // CHECK-NEXT: } +#pragma acc atomic update + f--; + + // CHECK-NEXT: acc.atomic.update %[[X_ALLOCA]] : !cir.ptr<!s32i> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !s32i{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[F_LOAD:.*]] = cir.load{{.*}} %[[F_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: %[[INT_TO_F:.*]] = cir.cast int_to_float %[[TEMP_LOAD]] : !s32i -> !cir.float + // CHECK-NEXT: %[[ADD:.*]] = cir.binop(add, %[[INT_TO_F]], %[[F_LOAD]]) : !cir.float + // CHECK-NEXT: %[[F_TO_INT:.*]] = cir.cast float_to_int %[[ADD]] : !cir.float -> !s32i + // CHECK-NEXT: cir.store{{.*}} %[[F_TO_INT]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !s32i + // CHECK-NEXT: } +#pragma acc atomic update + x += f; + + // CHECK-NEXT: acc.atomic.update %[[F_ALLOCA]] : !cir.ptr<!cir.float> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !cir.float{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !cir.float, !cir.ptr<!cir.float>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[Y_LOAD:.*]] = cir.load{{.*}} %[[Y_ALLOCA]] : !cir.ptr<!u32i>, !u32i + // CHECK-NEXT: %[[INT_TO_F:.*]] = cir.cast int_to_float %[[Y_LOAD]] : !u32i -> !cir.float + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: %[[DIV:.*]] = cir.binop(div, %[[TEMP_LOAD]], %[[INT_TO_F]]) : !cir.float + // CHECK-NEXT: cir.store{{.*}} %[[DIV]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !cir.float + // CHECK-NEXT: } +#pragma acc atomic update + f /= y; + + // CHECK-NEXT: acc.atomic.update %[[Y_ALLOCA]] : !cir.ptr<!u32i> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !u32i{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !u32i, !cir.ptr<!u32i>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !u32i, !cir.ptr<!u32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!u32i>, !u32i + // CHECK-NEXT: %[[CALL:.*]] = cir.call {{.*}}(%[[OPS_ALLOCA]]) : (!cir.ptr<!rec_HasOps>) -> !s32i + // CHECK-NEXT: %[[CALL_CAST:.*]] = cir.cast integral %[[CALL]] : !s32i -> !u32i + // CHECK-NEXT: %[[MUL:.*]] = cir.binop(mul, %[[TEMP_LOAD]], %[[CALL_CAST]]) : !u32i + // CHECK-NEXT: cir.store{{.*}} %[[MUL]], %[[TEMP_ALLOCA]] : !u32i, !cir.ptr<!u32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!u32i>, !u32i + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !u32i + // CHECK-NEXT: } + +#pragma acc atomic update + y = y * ops.thing(); + + // CHECK-NEXT: acc.atomic.update %[[X_ALLOCA]] : !cir.ptr<!s32i> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !s32i{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[CALL:.*]] = cir.call {{.*}}(%[[OPS_ALLOCA]]) : (!cir.ptr<!rec_HasOps>) -> !s32i + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: %[[OR:.*]] = cir.binop(or, %[[CALL]], %[[INT_TO_F]]) : !s32i + // CHECK-NEXT: cir.store{{.*}} %[[OR]], %[[TEMP_ALLOCA]] : !s32i, !cir.ptr<!s32i> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !s32i + // CHECK-NEXT: } +#pragma acc atomic update + x = ops.thing() | x; + + // CHECK-NEXT: %[[X_LOAD:.*]] = cir.load{{.*}} %[[X_ALLOCA]] : !cir.ptr<!s32i>, !s32i + // CHECK-NEXT: %[[BOOL_CAST:.*]] = cir.cast int_to_bool %[[X_LOAD]] : !s32i -> !cir.bool + // CHECK-NEXT: %[[X_CAST:.*]] = builtin.unrealized_conversion_cast %[[BOOL_CAST]] : !cir.bool to i1 + // CHECK-NEXT: acc.atomic.update if(%[[X_CAST]]) %[[F_ALLOCA]] : !cir.ptr<!cir.float> { + // CHECK-NEXT: ^bb0(%[[RECIPE_ARG:.*]]: !cir.float{{.*}}): + // CHECK-NEXT: %[[TEMP_ALLOCA:.*]] = cir.alloca !cir.float, !cir.ptr<!cir.float>, ["x_var", init] + // CHECK-NEXT: cir.store %[[RECIPE_ARG]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: %[[CALL:.*]] = cir.call {{.*}}(%[[OPS_ALLOCA]]) : (!cir.ptr<!rec_HasOps>) -> !cir.float + // CHECK-NEXT: %[[SUB:.*]] = cir.binop(sub, %[[TEMP_LOAD]], %[[CALL]]) : !cir.float + // CHECK-NEXT: cir.store{{.*}} %[[SUB]], %[[TEMP_ALLOCA]] : !cir.float, !cir.ptr<!cir.float> + // + // CHECK-NEXT: %[[TEMP_LOAD:.*]] = cir.load{{.*}} %[[TEMP_ALLOCA]] : !cir.ptr<!cir.float>, !cir.float + // CHECK-NEXT: acc.yield %[[TEMP_LOAD]] : !cir.float + // CHECK-NEXT: } +#pragma acc atomic update if (x) + f = f - ops; +} diff --git a/clang/test/CIR/CodeGenOpenACC/openacc-not-implemented.cpp b/clang/test/CIR/CodeGenOpenACC/openacc-not-implemented.cpp index 33e12fe..b4d76e1 100644 --- a/clang/test/CIR/CodeGenOpenACC/openacc-not-implemented.cpp +++ b/clang/test/CIR/CodeGenOpenACC/openacc-not-implemented.cpp @@ -3,8 +3,8 @@ void HelloWorld(int *A, int *B, int *C, int N) { // expected-error@+1{{ClangIR code gen Not Yet Implemented: OpenACC Atomic Construct}} -#pragma acc atomic - N = N + 1; +#pragma acc atomic capture + B = A += ++N; // expected-error@+1{{ClangIR code gen Not Yet Implemented: OpenACC Declare Construct}} #pragma acc declare create(A) diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e4m3.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e4m3.c new file mode 100644 index 0000000..d162f44 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e4m3.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e4m3_e4m3_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e4m3.e4m3.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e4m3_e4m3_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e4m3_e4m3_w4_u8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e5m2.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e5m2.c new file mode 100644 index 0000000..342af1e --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e4m3_e5m2.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e4m3_e5m2_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e4m3.e5m2.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e4m3_e5m2_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e4m3_e5m2_w4_u8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e4m3.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e4m3.c new file mode 100644 index 0000000..b8f58fe --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e4m3.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e5m2_e4m3_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e5m2.e4m3.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e5m2_e4m3_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e5m2_e4m3_w4_u8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e5m2.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e5m2.c new file mode 100644 index 0000000..7c2eb32 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_e5m2_e5m2.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e5m2_e5m2_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e5m2.e5m2.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e5m2_e5m2_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e5m2_e5m2_w4_u8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_f_f.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_f_f.c new file mode 100644 index 0000000..262bc0a --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_f_f.c @@ -0,0 +1,40 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a16f \ +// RUN: -target-feature +xsfmm32a32f -target-feature +xsfmm64a64f \ +// RUN: -target-feature +zvfhmin -target-feature +zve64d -disable-O0-optnone \ +// RUN: -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w2_f16m8( +// CHECK-RV64-SAME: <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x half> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv32f16(i64 0, <vscale x 32 x half> [[VS2]], <vscale x 32 x half> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 2) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w2_f16m8(vfloat16m8_t vs2, vfloat16m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w2_f16m8(0, vs2, vs1, tm, tn, tk); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w1_f32m8( +// CHECK-RV64-SAME: <vscale x 16 x float> [[VS2:%.*]], <vscale x 16 x float> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv16f32(i64 0, <vscale x 16 x float> [[VS2]], <vscale x 16 x float> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w1_f32m8(vfloat32m8_t vs2, vfloat32m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w1_f32m8(0, vs2, vs1, tm, tn, tk); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w1_f64m8( +// CHECK-RV64-SAME: <vscale x 8 x double> [[VS2:%.*]], <vscale x 8 x double> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv8f64(i64 0, <vscale x 8 x double> [[VS2]], <vscale x 8 x double> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w1_f64m8(vfloat64m8_t vs2, vfloat64m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w1_f64m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_s.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_s.c new file mode 100644 index 0000000..35c6756 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_s.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_s_s_w4_i8m8_i8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.s.s.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_s_s_w4_i8m8_i8m8(vint8m8_t vs2, vint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_s_s_w4_i8m8_i8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_u.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_u.c new file mode 100644 index 0000000..c142fcc --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_s_u.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_s_u_w4_i8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.s.u.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_s_u_w4_i8m8_u8m8(vint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_s_u_w4_i8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_s.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_s.c new file mode 100644 index 0000000..46350e5 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_s.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_u_s_w4_u8m8_i8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.u.s.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_u_s_w4_u8m8_i8m8(vuint8m8_t vs2, vint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_u_s_w4_u8m8_i8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_u.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_u.c new file mode 100644 index 0000000..de84db5 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_mm_u_u.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_u_u_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.u.u.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_u_u_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_u_u_w4_u8m8_u8m8(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte16.c new file mode 100644 index 0000000..2c23176 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte16.c @@ -0,0 +1,49 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_bf16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_bf16(size_t tss, __bf16 *base, size_t vl) { + return __riscv_sf_vlte16_bf16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_f16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_f16(size_t tss, _Float16 *base, size_t vl) { + return __riscv_sf_vlte16_f16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_i16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_i16(size_t tss, int16_t *base, size_t vl) { + return __riscv_sf_vlte16_i16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_u16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_u16(size_t tss, uint16_t *base, size_t vl) { + return __riscv_sf_vlte16_u16(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte32.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte32.c new file mode 100644 index 0000000..a0422cf --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte32.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_f32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_f32(size_t tss, float *base, size_t vl) { + return __riscv_sf_vlte32_f32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_i32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_i32(size_t tss, int32_t *base, size_t vl) { + return __riscv_sf_vlte32_i32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_u32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_u32(size_t tss, uint32_t *base, size_t vl) { + return __riscv_sf_vlte32_u32(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte64.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte64.c new file mode 100644 index 0000000..e8b9552 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte64.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_f64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_f64(size_t tss, double *base, size_t vl) { + return __riscv_sf_vlte64_f64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_i64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_i64(size_t tss, int64_t *base, size_t vl) { + return __riscv_sf_vlte64_i64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_u64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_u64(size_t tss, uint64_t *base, size_t vl) { + return __riscv_sf_vlte64_u64(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte8.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte8.c new file mode 100644 index 0000000..a86ccec --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vlte8.c @@ -0,0 +1,28 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte8_i8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte8_i8(size_t tss, int8_t *base, size_t vl) { + return __riscv_sf_vlte8_i8(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte8_u8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte8_u8(size_t tss, uint8_t *base, size_t vl) { + return __riscv_sf_vlte8_u8(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettk.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettk.c new file mode 100644 index 0000000..32b7bce8 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettk.c @@ -0,0 +1,99 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zve64x \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e8w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 0, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e8w1(size_t tn) { + return __riscv_sf_vsettk_e8w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e8w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 0, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e8w2(size_t tn) { + return __riscv_sf_vsettk_e8w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e8w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 0, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e8w4(size_t tn) { + return __riscv_sf_vsettk_e8w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e16w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 1, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e16w1(size_t tn) { + return __riscv_sf_vsettk_e16w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e16w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 1, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e16w2(size_t tn) { + return __riscv_sf_vsettk_e16w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e16w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 1, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e16w4(size_t tn) { + return __riscv_sf_vsettk_e16w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e32w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 2, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e32w1(size_t tn) { + return __riscv_sf_vsettk_e32w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e32w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 2, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e32w2(size_t tn) { + return __riscv_sf_vsettk_e32w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettk_e64w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettk.i64(i64 [[TN]], i64 3, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettk_e64w1(size_t tn) { + return __riscv_sf_vsettk_e64w1(tn); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettm.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettm.c new file mode 100644 index 0000000..0ce7c578 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettm.c @@ -0,0 +1,99 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zve64x \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e8w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 0, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e8w1(size_t tn) { + return __riscv_sf_vsettm_e8w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e8w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 0, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e8w2(size_t tn) { + return __riscv_sf_vsettm_e8w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e8w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 0, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e8w4(size_t tn) { + return __riscv_sf_vsettm_e8w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e16w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 1, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e16w1(size_t tn) { + return __riscv_sf_vsettm_e16w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e16w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 1, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e16w2(size_t tn) { + return __riscv_sf_vsettm_e16w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e16w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 1, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e16w4(size_t tn) { + return __riscv_sf_vsettm_e16w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e32w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 2, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e32w1(size_t tn) { + return __riscv_sf_vsettm_e32w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e32w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 2, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e32w2(size_t tn) { + return __riscv_sf_vsettm_e32w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettm_e64w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettm.i64(i64 [[TN]], i64 3, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettm_e64w1(size_t tn) { + return __riscv_sf_vsettm_e64w1(tn); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettn.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettn.c new file mode 100644 index 0000000..7b058f4 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettn.c @@ -0,0 +1,99 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zve64x \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e8w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e8w1(size_t tn) { + return __riscv_sf_vsettn_e8w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e8w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e8w2(size_t tn) { + return __riscv_sf_vsettn_e8w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e8w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e8w4(size_t tn) { + return __riscv_sf_vsettn_e8w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e16w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e16w1(size_t tn) { + return __riscv_sf_vsettn_e16w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e16w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e16w2(size_t tn) { + return __riscv_sf_vsettn_e16w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e16w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e16w4(size_t tn) { + return __riscv_sf_vsettn_e16w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e32w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 2, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e32w1(size_t tn) { + return __riscv_sf_vsettn_e32w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e32w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 2, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e32w2(size_t tn) { + return __riscv_sf_vsettn_e32w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettn_e64w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 3, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettn_e64w1(size_t tn) { + return __riscv_sf_vsettn_e64w1(tn); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettnt.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettnt.c new file mode 100644 index 0000000..29eaec3 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vsettnt.c @@ -0,0 +1,99 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zve64x \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e8w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e8w1(size_t tn) { + return __riscv_sf_vsettnt_e8w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e8w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e8w2(size_t tn) { + return __riscv_sf_vsettnt_e8w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e8w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 0, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e8w4(size_t tn) { + return __riscv_sf_vsettnt_e8w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e16w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e16w1(size_t tn) { + return __riscv_sf_vsettnt_e16w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e16w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e16w2(size_t tn) { + return __riscv_sf_vsettnt_e16w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e16w4( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 1, i64 3) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e16w4(size_t tn) { + return __riscv_sf_vsettnt_e16w4(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e32w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 2, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e32w1(size_t tn) { + return __riscv_sf_vsettnt_e32w1(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e32w2( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 2, i64 2) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e32w2(size_t tn) { + return __riscv_sf_vsettnt_e32w2(tn); +} + +// CHECK-RV64-LABEL: define dso_local i64 @test_sf_vsettnt_e64w1( +// CHECK-RV64-SAME: i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call i64 @llvm.riscv.sf.vsettnt.i64(i64 [[TN]], i64 3, i64 1) +// CHECK-RV64-NEXT: ret i64 [[TMP0]] +// +size_t test_sf_vsettnt_e64w1(size_t tn) { + return __riscv_sf_vsettnt_e64w1(tn); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste16.c new file mode 100644 index 0000000..bf50e7f --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste16.c @@ -0,0 +1,49 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_bf16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_bf16(size_t tss, __bf16 *base, size_t vl) { + return __riscv_sf_vste16_bf16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_f16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_f16(size_t tss, _Float16 *base, size_t vl) { + return __riscv_sf_vste16_f16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_i16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_i16(size_t tss, int16_t *base, size_t vl) { + return __riscv_sf_vste16_i16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_u16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_u16(size_t tss, uint16_t *base, size_t vl) { + return __riscv_sf_vste16_u16(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste32.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste32.c new file mode 100644 index 0000000..d1d7191 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste32.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_f32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_f32(size_t tss, float *base, size_t vl) { + return __riscv_sf_vste32_f32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_i32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_i32(size_t tss, int32_t *base, size_t vl) { + return __riscv_sf_vste32_i32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_u32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_u32(size_t tss, uint32_t *base, size_t vl) { + return __riscv_sf_vste32_u32(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste64.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste64.c new file mode 100644 index 0000000..4c7d4b1 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste64.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_f64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_f64(size_t tss, double *base, size_t vl) { + return __riscv_sf_vste64_f64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_i64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_i64(size_t tss, int64_t *base, size_t vl) { + return __riscv_sf_vste64_i64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_u64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_u64(size_t tss, uint64_t *base, size_t vl) { + return __riscv_sf_vste64_u64(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste8.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste8.c new file mode 100644 index 0000000..d03bc61 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vste8.c @@ -0,0 +1,28 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste8_i8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste8_i8(size_t tss, int8_t *base, size_t vl) { + return __riscv_sf_vste8_i8(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste8_u8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste8_u8(size_t tss, uint8_t *base, size_t vl) { + return __riscv_sf_vste8_u8(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtdiscard.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtdiscard.c new file mode 100644 index 0000000..7eef2c9 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtdiscard.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtdiscard( +// CHECK-RV64-SAME: ) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtdiscard() +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtdiscard() { + return __riscv_sf_vtdiscard(); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_t_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_t_v.c new file mode 100644 index 0000000..37ee503 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_t_v.c @@ -0,0 +1,130 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -target-feature +zve64d \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_bf16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32bf16.i64(i64 [[TSS]], <vscale x 32 x bfloat> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_bf16m8(size_t tss, vbfloat16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_bf16m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32f16.i64(i64 [[TSS]], <vscale x 32 x half> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f16m8(size_t tss, vfloat16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_f16m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16f32.i64(i64 [[TSS]], <vscale x 16 x float> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f32m8(size_t tss, vfloat32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_f32m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8f64.i64(i64 [[TSS]], <vscale x 8 x double> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f64m8(size_t tss, vfloat64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_f64m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv64i8.i64(i64 [[TSS]], <vscale x 64 x i8> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i8m8(size_t tss, vint8m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_i8m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32i16.i64(i64 [[TSS]], <vscale x 32 x i16> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i16m8(size_t tss, vint16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_i16m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16i32.i64(i64 [[TSS]], <vscale x 16 x i32> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i32m8(size_t tss, vint32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_i32m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8i64.i64(i64 [[TSS]], <vscale x 8 x i64> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i64m8(size_t tss, vint64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_i64m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv64i8.i64(i64 [[TSS]], <vscale x 64 x i8> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u8m8(size_t tss, vuint8m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_u8m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32i16.i64(i64 [[TSS]], <vscale x 32 x i16> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u16m8(size_t tss, vuint16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_u16m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16i32.i64(i64 [[TSS]], <vscale x 16 x i32> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u32m8(size_t tss, vuint32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_u32m8(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8i64.i64(i64 [[TSS]], <vscale x 8 x i64> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u64m8(size_t tss, vuint64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v_u64m8(tss, src, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_v_t.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_v_t.c new file mode 100644 index 0000000..d127cf4 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtmv_v_t.c @@ -0,0 +1,130 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -target-feature +zve64d \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_sf_vtmv_v_t_bf16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.sf.vtmv.v.t.nxv32bf16.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 32 x bfloat> [[TMP0]] +// +vbfloat16m8_t test_sf_vtmv_v_t_bf16m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_bf16m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 32 x half> @test_sf_vtmv_v_t_f16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 32 x half> @llvm.riscv.sf.vtmv.v.t.nxv32f16.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 32 x half> [[TMP0]] +// +vfloat16m8_t test_sf_vtmv_v_t_f16m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_f16m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 16 x float> @test_sf_vtmv_v_t_f32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 16 x float> @llvm.riscv.sf.vtmv.v.t.nxv16f32.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 16 x float> [[TMP0]] +// +vfloat32m8_t test_sf_vtmv_v_t_f32m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_f32m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 8 x double> @test_sf_vtmv_v_t_f64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 8 x double> @llvm.riscv.sf.vtmv.v.t.nxv8f64.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 8 x double> [[TMP0]] +// +vfloat64m8_t test_sf_vtmv_v_t_f64m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_f64m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 64 x i8> @test_sf_vtmv_v_t_i8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.sf.vtmv.v.t.nxv64i8.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 64 x i8> [[TMP0]] +// +vint8m8_t test_sf_vtmv_v_t_i8m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_i8m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_sf_vtmv_v_t_i16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.sf.vtmv.v.t.nxv32i16.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 32 x i16> [[TMP0]] +// +vint16m8_t test_sf_vtmv_v_t_i16m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_i16m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vtmv_v_t_i32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vtmv.v.t.nxv16i32.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 16 x i32> [[TMP0]] +// +vint32m8_t test_sf_vtmv_v_t_i32m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_i32m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i64> @test_sf_vtmv_v_t_i64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.sf.vtmv.v.t.nxv8i64.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 8 x i64> [[TMP0]] +// +vint64m8_t test_sf_vtmv_v_t_i64m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_i64m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 64 x i8> @test_sf_vtmv_v_t_u8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 64 x i8> @llvm.riscv.sf.vtmv.v.t.nxv64i8.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 64 x i8> [[TMP0]] +// +vuint8m8_t test_sf_vtmv_v_t_u8m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_u8m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_sf_vtmv_v_t_u16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 32 x i16> @llvm.riscv.sf.vtmv.v.t.nxv32i16.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 32 x i16> [[TMP0]] +// +vuint16m8_t test_sf_vtmv_v_t_u16m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_u16m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_sf_vtmv_v_t_u32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.sf.vtmv.v.t.nxv16i32.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 16 x i32> [[TMP0]] +// +vuint32m8_t test_sf_vtmv_v_t_u32m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_u32m8(tss, vl); +} + +// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i64> @test_sf_vtmv_v_t_u64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: [[TMP0:%.*]] = call <vscale x 8 x i64> @llvm.riscv.sf.vtmv.v.t.nxv8i64.i64(i64 [[TSS]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret <vscale x 8 x i64> [[TMP0]] +// +vuint64m8_t test_sf_vtmv_v_t_u64m8(size_t tss, size_t vl) { + return __riscv_sf_vtmv_v_t_u64m8(tss, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtzero_t.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtzero_t.c new file mode 100644 index 0000000..1e4de48 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/non-overloaded/sf_vtzero_t.c @@ -0,0 +1,99 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zve64x \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e8w1( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 3, i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e8w1(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e8w1(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e8w2( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 3, i64 2) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e8w2(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e8w2(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e8w4( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 3, i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e8w4(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e8w4(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e16w1( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 4, i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e16w1(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e16w1(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e16w2( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 4, i64 2) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e16w2(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e16w2(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e16w4( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 4, i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e16w4(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e16w4(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e32w1( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 5, i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e32w1(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e32w1(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e32w2( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 5, i64 2) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e32w2(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e32w2(0, tm, tn); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtzero_t_e64w1( +// CHECK-RV64-SAME: i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtzero.t.i64(i64 0, i64 [[TM]], i64 [[TN]], i64 6, i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtzero_t_e64w1(size_t tm, size_t tn) { + return __riscv_sf_vtzero_t_e64w1(0, tm, tn); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e4m3.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e4m3.c new file mode 100644 index 0000000..2f6c4dc --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e4m3.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e4m3_e4m3_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e4m3.e4m3.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e4m3_e4m3_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e4m3_e4m3(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e5m2.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e5m2.c new file mode 100644 index 0000000..40ae780 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e4m3_e5m2.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e4m3_e5m2_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e4m3.e5m2.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e4m3_e5m2_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e4m3_e5m2(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e4m3.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e4m3.c new file mode 100644 index 0000000..f4f024c --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e4m3.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e5m2_e4m3_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e5m2.e4m3.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e5m2_e4m3_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e5m2_e4m3(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e5m2.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e5m2.c new file mode 100644 index 0000000..01399d5 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_e5m2_e5m2.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8f \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_e5m2_e5m2_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.e5m2.e5m2.i64.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_e5m2_e5m2_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_e5m2_e5m2(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_f_f.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_f_f.c new file mode 100644 index 0000000..2371e4e --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_f_f.c @@ -0,0 +1,40 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a16f \ +// RUN: -target-feature +xsfmm32a32f -target-feature +xsfmm64a64f \ +// RUN: -target-feature +zvfhmin -target-feature +zve64d -disable-O0-optnone \ +// RUN: -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w2_f16m8( +// CHECK-RV64-SAME: <vscale x 32 x half> [[VS2:%.*]], <vscale x 32 x half> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv32f16(i64 0, <vscale x 32 x half> [[VS2]], <vscale x 32 x half> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 2) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w2_f16m8(vfloat16m8_t vs2, vfloat16m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w2(0, vs2, vs1, tm, tn, tk); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w1_f32m8( +// CHECK-RV64-SAME: <vscale x 16 x float> [[VS2:%.*]], <vscale x 16 x float> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv16f32(i64 0, <vscale x 16 x float> [[VS2]], <vscale x 16 x float> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w1_f32m8(vfloat32m8_t vs2, vfloat32m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w1(0, vs2, vs1, tm, tn, tk); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_f_f_w1_f64m8( +// CHECK-RV64-SAME: <vscale x 8 x double> [[VS2:%.*]], <vscale x 8 x double> [[VS1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.f.f.i64.nxv8f64(i64 0, <vscale x 8 x double> [[VS2]], <vscale x 8 x double> [[VS1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 1) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_f_f_w1_f64m8(vfloat64m8_t vs2, vfloat64m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_f_f_w1(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_s.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_s.c new file mode 100644 index 0000000..2d34f7d --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_s.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_s_s_w4_i8m8_i8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.s.s.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_s_s_w4_i8m8_i8m8(vint8m8_t vs2, vint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_s_s(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_u.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_u.c new file mode 100644 index 0000000..1f9bc33 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_s_u.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_s_u_w4_i8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.s.u.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_s_u_w4_i8m8_u8m8(vint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_s_u(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_s.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_s.c new file mode 100644 index 0000000..2c6d538 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_s.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_u_s_w4_u8m8_i8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.u.s.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_u_s_w4_u8m8_i8m8(vuint8m8_t vs2, vint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_u_s(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_u.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_u.c new file mode 100644 index 0000000..bb1eaf1 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_mm_u_u.c @@ -0,0 +1,18 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmm32a8i \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_mm_u_u_w4_u8m8_u8m8( +// CHECK-RV64-SAME: <vscale x 64 x i8> [[vs2:%.*]], <vscale x 64 x i8> [[vs1:%.*]], i64 noundef [[TM:%.*]], i64 noundef [[TN:%.*]], i64 noundef [[TK:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.mm.u.u.i64.nxv64i8.nxv64i8(i64 0, <vscale x 64 x i8> [[vs2]], <vscale x 64 x i8> [[vs1]], i64 [[TM]], i64 [[TN]], i64 [[TK]], i64 4) +// CHECK-RV64-NEXT: ret void +// +void test_sf_mm_u_u_w4_u8m8_u8m8(vuint8m8_t vs2, vuint8m8_t vs1, size_t tm, size_t tn, size_t tk) { + return __riscv_sf_mm_u_u(0, vs2, vs1, tm, tn, tk); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte16.c new file mode 100644 index 0000000..e199c1f --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte16.c @@ -0,0 +1,49 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_bf16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_bf16(size_t tss, __bf16 *base, size_t vl) { + return __riscv_sf_vlte16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_f16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_f16(size_t tss, _Float16 *base, size_t vl) { + return __riscv_sf_vlte16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_i16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_i16(size_t tss, int16_t *base, size_t vl) { + return __riscv_sf_vlte16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte16_u16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte16_u16(size_t tss, uint16_t *base, size_t vl) { + return __riscv_sf_vlte16(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte32.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte32.c new file mode 100644 index 0000000..388884d --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte32.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_f32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_f32(size_t tss, float *base, size_t vl) { + return __riscv_sf_vlte32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_i32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_i32(size_t tss, int32_t *base, size_t vl) { + return __riscv_sf_vlte32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte32_u32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte32_u32(size_t tss, uint32_t *base, size_t vl) { + return __riscv_sf_vlte32(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte64.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte64.c new file mode 100644 index 0000000..80d7542 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte64.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_f64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_f64(size_t tss, double *base, size_t vl) { + return __riscv_sf_vlte64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_i64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_i64(size_t tss, int64_t *base, size_t vl) { + return __riscv_sf_vlte64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte64_u64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte64_u64(size_t tss, uint64_t *base, size_t vl) { + return __riscv_sf_vlte64(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte8.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte8.c new file mode 100644 index 0000000..c29bf933 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vlte8.c @@ -0,0 +1,28 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte8_i8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte8_i8(size_t tss, int8_t *base, size_t vl) { + return __riscv_sf_vlte8(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vlte8_u8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vlte8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vlte8_u8(size_t tss, uint8_t *base, size_t vl) { + return __riscv_sf_vlte8(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste16.c new file mode 100644 index 0000000..e1fff6c --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste16.c @@ -0,0 +1,49 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_bf16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_bf16(size_t tss, __bf16 *base, size_t vl) { + return __riscv_sf_vste16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_f16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_f16(size_t tss, _Float16 *base, size_t vl) { + return __riscv_sf_vste16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_i16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_i16(size_t tss, int16_t *base, size_t vl) { + return __riscv_sf_vste16(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste16_u16( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste16.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste16_u16(size_t tss, uint16_t *base, size_t vl) { + return __riscv_sf_vste16(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste32.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste32.c new file mode 100644 index 0000000..0c3bc4c --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste32.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_f32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_f32(size_t tss, float *base, size_t vl) { + return __riscv_sf_vste32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_i32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_i32(size_t tss, int32_t *base, size_t vl) { + return __riscv_sf_vste32(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste32_u32( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste32.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste32_u32(size_t tss, uint32_t *base, size_t vl) { + return __riscv_sf_vste32(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste64.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste64.c new file mode 100644 index 0000000..0a56807 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste64.c @@ -0,0 +1,38 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_f64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_f64(size_t tss, double *base, size_t vl) { + return __riscv_sf_vste64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_i64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_i64(size_t tss, int64_t *base, size_t vl) { + return __riscv_sf_vste64(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste64_u64( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste64.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste64_u64(size_t tss, uint64_t *base, size_t vl) { + return __riscv_sf_vste64(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste8.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste8.c new file mode 100644 index 0000000..3115945 --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vste8.c @@ -0,0 +1,28 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste8_i8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste8_i8(size_t tss, int8_t *base, size_t vl) { + return __riscv_sf_vste8(tss, base, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vste8_u8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], ptr noundef [[BASE:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vste8.i64(i64 [[TSS]], ptr [[BASE]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vste8_u8(size_t tss, uint8_t *base, size_t vl) { + return __riscv_sf_vste8(tss, base, vl); +} + diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vtmv_t_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vtmv_t_v.c new file mode 100644 index 0000000..2461c39c --- /dev/null +++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-sifive/non-policy/overloaded/sf_vtmv_t_v.c @@ -0,0 +1,130 @@ +// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4 +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +xsfmmbase \ +// RUN: -target-feature +zvfhmin -target-feature +zvfbfmin \ +// RUN: -target-feature +zve64d \ +// RUN: -disable-O0-optnone -emit-llvm %s -o - | \ +// RUN: opt -S -passes=mem2reg | FileCheck --check-prefix=CHECK-RV64 %s + +#include <sifive_vector.h> + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_bf16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32bf16.i64(i64 [[TSS]], <vscale x 32 x bfloat> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_bf16m8(size_t tss, vbfloat16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x half> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32f16.i64(i64 [[TSS]], <vscale x 32 x half> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f16m8(size_t tss, vfloat16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x float> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16f32.i64(i64 [[TSS]], <vscale x 16 x float> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f32m8(size_t tss, vfloat32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_f64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x double> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8f64.i64(i64 [[TSS]], <vscale x 8 x double> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_f64m8(size_t tss, vfloat64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv64i8.i64(i64 [[TSS]], <vscale x 64 x i8> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i8m8(size_t tss, vint8m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32i16.i64(i64 [[TSS]], <vscale x 32 x i16> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i16m8(size_t tss, vint16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16i32.i64(i64 [[TSS]], <vscale x 16 x i32> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i32m8(size_t tss, vint32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_i64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8i64.i64(i64 [[TSS]], <vscale x 8 x i64> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_i64m8(size_t tss, vint64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u8m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 64 x i8> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv64i8.i64(i64 [[TSS]], <vscale x 64 x i8> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u8m8(size_t tss, vuint8m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u16m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 32 x i16> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv32i16.i64(i64 [[TSS]], <vscale x 32 x i16> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u16m8(size_t tss, vuint16m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u32m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 16 x i32> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv16i32.i64(i64 [[TSS]], <vscale x 16 x i32> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u32m8(size_t tss, vuint32m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + +// CHECK-RV64-LABEL: define dso_local void @test_sf_vtmv_t_v_u64m8( +// CHECK-RV64-SAME: i64 noundef [[TSS:%.*]], <vscale x 8 x i64> [[SRC:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] { +// CHECK-RV64-NEXT: entry: +// CHECK-RV64-NEXT: call void @llvm.riscv.sf.vtmv.t.v.nxv8i64.i64(i64 [[TSS]], <vscale x 8 x i64> [[SRC]], i64 [[VL]]) +// CHECK-RV64-NEXT: ret void +// +void test_sf_vtmv_t_v_u64m8(size_t tss, vuint64m8_t src, size_t vl) { + return __riscv_sf_vtmv_t_v(tss, src, vl); +} + diff --git a/clang/test/CodeGen/X86/avx-builtins.c b/clang/test/CodeGen/X86/avx-builtins.c index 26aed772..737febb 100644 --- a/clang/test/CodeGen/X86/avx-builtins.c +++ b/clang/test/CodeGen/X86/avx-builtins.c @@ -1891,12 +1891,16 @@ __m256d test_mm256_shuffle_pd(__m256d A, __m256d B) { return _mm256_shuffle_pd(A, B, 0); } +TEST_CONSTEXPR((match_m256d(_mm256_shuffle_pd(((__m256d)(__v4df){1.0, 2.0, 3.0, 4.0}), ((__m256d)(__v4df){5.0, 6.0, 7.0, 8.0}), 15), 2.0, 6.0, 4.0, 8.0))); + __m256 test_mm256_shuffle_ps(__m256 A, __m256 B) { // CHECK-LABEL: test_mm256_shuffle_ps // CHECK: shufflevector <8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x i32> <i32 0, i32 0, i32 8, i32 8, i32 4, i32 4, i32 12, i32 12> return _mm256_shuffle_ps(A, B, 0); } +TEST_CONSTEXPR((match_m256(_mm256_shuffle_ps(((__m256)(__v8sf){1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f}), ((__m256)(__v8sf){9.0f, 10.0f, 11.0f, 12.0f, 13.0f, 14.0f, 15.0f, 16.0f}), 4), 1.0f, 2.0f, 9.0f, 9.0f, 5.0f, 6.0f, 13.0f, 13.0f))); + __m256d test_mm256_sqrt_pd(__m256d A) { // CHECK-LABEL: test_mm256_sqrt_pd // CHECK: call {{.*}}<4 x double> @llvm.sqrt.v4f64(<4 x double> %{{.*}}) diff --git a/clang/test/CodeGen/X86/avx2-builtins.c b/clang/test/CodeGen/X86/avx2-builtins.c index 5c52d84..de4cb2f 100644 --- a/clang/test/CodeGen/X86/avx2-builtins.c +++ b/clang/test/CodeGen/X86/avx2-builtins.c @@ -1236,6 +1236,8 @@ __m256i test_mm256_slli_si256(__m256i a) { // CHECK: shufflevector <32 x i8> zeroinitializer, <32 x i8> %{{.*}}, <32 x i32> <i32 13, i32 14, i32 15, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 29, i32 30, i32 31, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60> return _mm256_slli_si256(a, 3); } +TEST_CONSTEXPR(match_v32qi(_mm256_slli_si256(((__m256i)(__v32qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}), 3), 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 0, 0, 0, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)); +TEST_CONSTEXPR(match_v32qi(_mm256_slli_si256(((__m256i)(__v32qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); __m128i test_mm_sllv_epi32(__m128i a, __m128i b) { // CHECK-LABEL: test_mm_sllv_epi32 @@ -1379,6 +1381,8 @@ __m256i test_mm256_srli_si256(__m256i a) { // CHECK: shufflevector <32 x i8> %{{.*}}, <32 x i8> zeroinitializer, <32 x i32> <i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 32, i32 33, i32 34, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 48, i32 49, i32 50> return _mm256_srli_si256(a, 3); } +TEST_CONSTEXPR(match_v32qi(_mm256_srli_si256(((__m256i)(__v32qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}), 3), 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 0, 0, 0)); +TEST_CONSTEXPR(match_v32qi(_mm256_srli_si256(((__m256i)(__v32qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); __m128i test_mm_srlv_epi32(__m128i a, __m128i b) { // CHECK-LABEL: test_mm_srlv_epi32 diff --git a/clang/test/CodeGen/X86/avx512bw-builtins.c b/clang/test/CodeGen/X86/avx512bw-builtins.c index d07e40a..be2cd48 100644 --- a/clang/test/CodeGen/X86/avx512bw-builtins.c +++ b/clang/test/CodeGen/X86/avx512bw-builtins.c @@ -2045,6 +2045,8 @@ __m512i test_mm512_bslli_epi128(__m512i __A) { // CHECK: shufflevector <64 x i8> zeroinitializer, <64 x i8> %{{.*}}, <64 x i32> <i32 11, i32 12, i32 13, i32 14, i32 15, i32 64, i32 65, i32 66, i32 67, i32 68, i32 69, i32 70, i32 71, i32 72, i32 73, i32 74, i32 27, i32 28, i32 29, i32 30, i32 31, i32 80, i32 81, i32 82, i32 83, i32 84, i32 85, i32 86, i32 87, i32 88, i32 89, i32 90, i32 43, i32 44, i32 45, i32 46, i32 47, i32 96, i32 97, i32 98, i32 99, i32 100, i32 101, i32 102, i32 103, i32 104, i32 105, i32 106, i32 59, i32 60, i32 61, i32 62, i32 63, i32 112, i32 113, i32 114, i32 115, i32 116, i32 117, i32 118, i32 119, i32 120, i32 121, i32 122> return _mm512_bslli_epi128(__A, 5); } +TEST_CONSTEXPR(match_v64qi(_mm512_bslli_epi128(((__m512i)(__v64qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64}), 5), 0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 0, 0, 0, 0, 0, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 0, 0, 0, 0, 0, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 0, 0, 0, 0, 0, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59)); +TEST_CONSTEXPR(match_v64qi(_mm512_bslli_epi128(((__m512i)(__v64qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); __m512i test_mm512_srlv_epi16(__m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_srlv_epi16 @@ -2223,6 +2225,9 @@ __m512i test_mm512_bsrli_epi128(__m512i __A) { // CHECK: shufflevector <64 x i8> %{{.*}}, <64 x i8> zeroinitializer, <64 x i32> <i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 64, i32 65, i32 66, i32 67, i32 68, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 80, i32 81, i32 82, i32 83, i32 84, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 96, i32 97, i32 98, i32 99, i32 100, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63, i32 112, i32 113, i32 114, i32 115, i32 116> return _mm512_bsrli_epi128(__A, 5); } +TEST_CONSTEXPR(match_v64qi(_mm512_bsrli_epi128(((__m512i)(__v64qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64}), 5), 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 0, 0, 0, 0, 0, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 0, 0, 0, 0, 0, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 0, 0, 0, 0, 0, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 0, 0, 0, 0, 0)); +TEST_CONSTEXPR(match_v64qi(_mm512_bsrli_epi128(((__m512i)(__v64qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); + __m512i test_mm512_mask_mov_epi16(__m512i __W, __mmask32 __U, __m512i __A) { // CHECK-LABEL: test_mm512_mask_mov_epi16 // CHECK: select <32 x i1> %{{.*}}, <32 x i16> %{{.*}}, <32 x i16> %{{.*}} diff --git a/clang/test/CodeGen/X86/avx512f-builtins.c b/clang/test/CodeGen/X86/avx512f-builtins.c index 122dadd..6959937 100644 --- a/clang/test/CodeGen/X86/avx512f-builtins.c +++ b/clang/test/CodeGen/X86/avx512f-builtins.c @@ -6741,9 +6741,13 @@ __m512 test_mm512_maskz_shuffle_ps(__mmask16 __U, __m512 __M, __m512 __V) { // CHECK-LABEL: test_mm512_maskz_shuffle_ps // CHECK: shufflevector <16 x float> %{{.*}}, <16 x float> %{{.*}}, <16 x i32> <i32 0, i32 1, i32 16, i32 16, i32 4, i32 5, i32 20, i32 20, i32 8, i32 9, i32 24, i32 24, i32 12, i32 13, i32 28, i32 28> // CHECK: select <16 x i1> %{{.*}}, <16 x float> %{{.*}}, <16 x float> %{{.*}} - return _mm512_maskz_shuffle_ps(__U, __M, __V, 4); + return _mm512_maskz_shuffle_ps(__U, __M, __V, 4); } +TEST_CONSTEXPR((match_m512(_mm512_shuffle_ps(((__m512)(__v16sf){1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f, 9.0f, 10.0f, 11.0f, 12.0f, 13.0f, 14.0f, 15.0f, 16.0f}), ((__m512)(__v16sf){17.0f, 18.0f, 19.0f, 20.0f, 21.0f, 22.0f, 23.0f, 24.0f, 25.0f, 26.0f, 27.0f, 28.0f, 29.0f, 30.0f, 31.0f, 32.0f}), 4), 1.0f, 2.0f, 17.0f, 17.0f, 5.0f, 6.0f, 21.0f, 21.0f, 9.0f, 10.0f, 25.0f, 25.0f, 13.0f, 14.0f, 29.0f, 29.0f))); +TEST_CONSTEXPR((match_m512d(_mm512_shuffle_pd(((__m512d)(__v8df){1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0}), ((__m512d)(__v8df){9.0, 10.0, 11.0, 12.0, 13.0, 14.0, 15.0, 16.0}), 48), 1.0, 9.0, 3.0, 11.0, 6.0, 14.0, 7.0, 15.0))); +TEST_CONSTEXPR((match_m512d(_mm512_maskz_shuffle_pd(0xFF, ((__m512d)(__v8df){1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0}), ((__m512d)(__v8df){9.0, 10.0, 11.0, 12.0, 13.0, 14.0, 15.0, 16.0}), 48), 1.0, 9.0, 3.0, 11.0, 6.0, 14.0, 7.0, 15.0))); + __m128d test_mm_sqrt_round_sd(__m128d __A, __m128d __B) { // CHECK-LABEL: test_mm_sqrt_round_sd // CHECK: call {{.*}}<2 x double> @llvm.x86.avx512.mask.sqrt.sd(<2 x double> %{{.*}}, <2 x double> %{{.*}}, <2 x double> %{{.*}}, i8 -1, i32 11) diff --git a/clang/test/CodeGen/X86/avx512vl-builtins.c b/clang/test/CodeGen/X86/avx512vl-builtins.c index 34db764..33c4397 100644 --- a/clang/test/CodeGen/X86/avx512vl-builtins.c +++ b/clang/test/CodeGen/X86/avx512vl-builtins.c @@ -8933,9 +8933,14 @@ __m256 test_mm256_maskz_shuffle_ps(__mmask8 __U, __m256 __A, __m256 __B) { // CHECK-LABEL: test_mm256_maskz_shuffle_ps // CHECK: shufflevector <8 x float> %{{.*}}, <8 x float> %{{.*}}, <8 x i32> <i32 0, i32 1, i32 8, i32 8, i32 4, i32 5, i32 12, i32 12> // CHECK: select <8 x i1> %{{.*}}, <8 x float> %{{.*}}, <8 x float> %{{.*}} - return _mm256_maskz_shuffle_ps(__U, __A, __B, 4); + return _mm256_maskz_shuffle_ps(__U, __A, __B, 4); } +TEST_CONSTEXPR((match_m128d(_mm_maskz_shuffle_pd(0x3, ((__m128d)(__v2df){1.0, 2.0}), ((__m128d)(__v2df){3.0, 4.0}), 3), 2.0, 4.0))); +TEST_CONSTEXPR((match_m256d(_mm256_maskz_shuffle_pd(0xF, ((__m256d)(__v4df){1.0, 2.0, 3.0, 4.0}), ((__m256d)(__v4df){5.0, 6.0, 7.0, 8.0}), 15), 2.0, 6.0, 4.0, 8.0))); +TEST_CONSTEXPR((match_m128(_mm_maskz_shuffle_ps(0xF, ((__m128)(__v4sf){1.0f, 2.0f, 3.0f, 4.0f}), ((__m128)(__v4sf){5.0f, 6.0f, 7.0f, 8.0f}), 4), 1.0f, 2.0f, 5.0f, 5.0f))); +TEST_CONSTEXPR((match_m256(_mm256_maskz_shuffle_ps(0xFF, ((__m256)(__v8sf){1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f, 7.0f, 8.0f}), ((__m256)(__v8sf){9.0f, 10.0f, 11.0f, 12.0f, 13.0f, 14.0f, 15.0f, 16.0f}), 4), 1.0f, 2.0f, 9.0f, 9.0f, 5.0f, 6.0f, 13.0f, 13.0f))); + __m128d test_mm_rsqrt14_pd(__m128d __A) { // CHECK-LABEL: test_mm_rsqrt14_pd // CHECK: @llvm.x86.avx512.rsqrt14.pd.128 diff --git a/clang/test/CodeGen/X86/sse-builtins.c b/clang/test/CodeGen/X86/sse-builtins.c index f5c1d00..6c5297e 100644 --- a/clang/test/CodeGen/X86/sse-builtins.c +++ b/clang/test/CodeGen/X86/sse-builtins.c @@ -738,6 +738,11 @@ __m128 test_mm_shuffle_ps(__m128 A, __m128 B) { return _mm_shuffle_ps(A, B, 0); } +TEST_CONSTEXPR((match_m128(_mm_shuffle_ps(((__m128)(__v4sf){1.0f, 2.0f, 3.0f, 4.0f}), ((__m128)(__v4sf){5.0f, 6.0f, 7.0f, 8.0f}), 4), 1.0f, 2.0f, 5.0f, 5.0f))); +TEST_CONSTEXPR((match_m128(_mm_shuffle_ps(((__m128)(__v4sf){1.0f, 2.0f, 3.0f, 4.0f}), ((__m128)(__v4sf){5.0f, 6.0f, 7.0f, 8.0f}), 0), 1.0f, 1.0f, 5.0f, 5.0f))); +TEST_CONSTEXPR((match_m128(_mm_shuffle_ps(((__m128)(__v4sf){1.0f, 2.0f, 3.0f, 4.0f}), ((__m128)(__v4sf){5.0f, 6.0f, 7.0f, 8.0f}), 255), 4.0f, 4.0f, 8.0f, 8.0f))); +TEST_CONSTEXPR((match_m128(_mm_shuffle_ps(((__m128)(__v4sf){1.0f, 2.0f, 3.0f, 4.0f}), ((__m128)(__v4sf){5.0f, 6.0f, 7.0f, 8.0f}), 27), 4.0f, 3.0f, 6.0f, 5.0f))); + __m128 test_mm_sqrt_ps(__m128 x) { // CHECK-LABEL: test_mm_sqrt_ps // CHECK: call {{.*}}<4 x float> @llvm.sqrt.v4f32(<4 x float> {{.*}}) diff --git a/clang/test/CodeGen/X86/sse2-builtins.c b/clang/test/CodeGen/X86/sse2-builtins.c index 8e4fb86..379ae48 100644 --- a/clang/test/CodeGen/X86/sse2-builtins.c +++ b/clang/test/CodeGen/X86/sse2-builtins.c @@ -1314,6 +1314,11 @@ __m128d test_mm_shuffle_pd(__m128d A, __m128d B) { return _mm_shuffle_pd(A, B, 1); } +TEST_CONSTEXPR((match_m128d(_mm_shuffle_pd(((__m128d)(__v2df){1.0, 2.0}), ((__m128d)(__v2df){3.0, 4.0}), 3), 2.0, 4.0))); +TEST_CONSTEXPR((match_m128d(_mm_shuffle_pd(((__m128d)(__v2df){1.0, 2.0}), ((__m128d)(__v2df){3.0, 4.0}), 0), 1.0, 3.0))); +TEST_CONSTEXPR((match_m128d(_mm_shuffle_pd(((__m128d)(__v2df){1.0, 2.0}), ((__m128d)(__v2df){3.0, 4.0}), 1), 2.0, 3.0))); +TEST_CONSTEXPR((match_m128d(_mm_shuffle_pd(((__m128d)(__v2df){1.0, 2.0}), ((__m128d)(__v2df){3.0, 4.0}), 2), 1.0, 4.0))); + __m128i test_mm_shufflehi_epi16(__m128i A) { // CHECK-LABEL: test_mm_shufflehi_epi16 // CHECK: shufflevector <8 x i16> %{{.*}}, <8 x i16> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 4, i32 4, i32 4> @@ -1418,6 +1423,8 @@ __m128i test_mm_slli_si128(__m128i A) { // CHECK: shufflevector <16 x i8> zeroinitializer, <16 x i8> %{{.*}}, <16 x i32> <i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26> return _mm_slli_si128(A, 5); } +TEST_CONSTEXPR(match_v16qi(_mm_slli_si128(((__m128i)(__v16qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}), 5), 0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11)); +TEST_CONSTEXPR(match_v16qi(_mm_slli_si128(((__m128i)(__v16qi){1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); __m128i test_mm_slli_si128_2(__m128i A) { // CHECK-LABEL: test_mm_slli_si128_2 @@ -1570,6 +1577,8 @@ __m128i test_mm_srli_si128(__m128i A) { // CHECK: shufflevector <16 x i8> %{{.*}}, <16 x i8> zeroinitializer, <16 x i32> <i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20> return _mm_srli_si128(A, 5); } +TEST_CONSTEXPR(match_v16qi(_mm_srli_si128(((__m128i)(__v16qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}), 5), 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 0, 0, 0, 0, 0)); +TEST_CONSTEXPR(match_v16qi(_mm_srli_si128(((__m128i)(__v16qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}), 16), 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)); __m128i test_mm_srli_si128_2(__m128i A) { // CHECK-LABEL: test_mm_srli_si128_2 diff --git a/clang/test/Driver/aarch64-ptrauth.c b/clang/test/Driver/aarch64-ptrauth.c index 5bb963a..b080a77 100644 --- a/clang/test/Driver/aarch64-ptrauth.c +++ b/clang/test/Driver/aarch64-ptrauth.c @@ -22,25 +22,39 @@ // RUN: %clang -### -c --target=aarch64-linux -mabi=pauthtest %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI1 // RUN: %clang -### -c --target=aarch64-linux-pauthtest %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI1 // PAUTHABI1: "-cc1"{{.*}} "-triple" "aarch64-unknown-linux-pauthtest" -// PAUTHABI1-SAME: "-target-abi" "pauthtest" // PAUTHABI1-SAME: "-fptrauth-intrinsics" "-fptrauth-calls" "-fptrauth-returns" "-fptrauth-auth-traps" "-fptrauth-vtable-pointer-address-discrimination" "-fptrauth-vtable-pointer-type-discrimination" "-fptrauth-type-info-vtable-pointer-discrimination" "-fptrauth-indirect-gotos" "-fptrauth-init-fini" "-fptrauth-init-fini-address-discrimination" "-faarch64-jump-table-hardening" +// PAUTHABI1-SAME: "-target-abi" "pauthtest" +// PAUTHABI1-NOT: "-fptrauth-function-pointer-type-discrimination" -// RUN: %clang -### -c --target=aarch64 -mabi=pauthtest -fno-ptrauth-intrinsics \ +// RUN: %clang -### -c --target=aarch64-linux -mabi=pauthtest -fno-ptrauth-intrinsics \ // RUN: -fno-ptrauth-calls -fno-ptrauth-returns -fno-ptrauth-auth-traps \ // RUN: -fno-ptrauth-vtable-pointer-address-discrimination -fno-ptrauth-vtable-pointer-type-discrimination \ // RUN: -fno-ptrauth-type-info-vtable-pointer-discrimination -fno-ptrauth-indirect-gotos \ // RUN: -fno-ptrauth-init-fini -fno-ptrauth-init-fini-address-discrimination \ // RUN: -fno-aarch64-jump-table-hardening %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI2 -// RUN: %clang -### -c --target=aarch64-pauthtest -fno-ptrauth-intrinsics \ +// RUN: %clang -### -c --target=aarch64-linux-pauthtest -fno-ptrauth-intrinsics \ // RUN: -fno-ptrauth-calls -fno-ptrauth-returns -fno-ptrauth-auth-traps \ // RUN: -fno-ptrauth-vtable-pointer-address-discrimination -fno-ptrauth-vtable-pointer-type-discrimination \ // RUN: -fno-ptrauth-type-info-vtable-pointer-discrimination -fno-ptrauth-indirect-gotos \ // RUN: -fno-ptrauth-init-fini -fno-ptrauth-init-fini-address-discrimination \ // RUN: -fno-aarch64-jump-table-hardening %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI2 -// PAUTHABI2: "-cc1" -// PAUTHABI2-NOT: "-fptrauth- + +//// Non-linux OS: pauthtest ABI has no effect in terms of passing ptrauth cc1 flags. +//// An error about unsupported ABI will be emitted later in pipeline (see ERR2 below) +// RUN: %clang -### -c --target=aarch64 -mabi=pauthtest %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI2 + +// PAUTHABI2: "-cc1" +// PAUTHABI2-SAME: "-target-abi" "pauthtest" +// PAUTHABI2-NOT: "-fptrauth- // PAUTHABI2-NOT: "-faarch64-jump-table-hardening" +//// Non-linux OS: pauthtest environment does not correspond to pauthtest ABI; aapcs is the default. +// RUN: %clang -### -c --target=aarch64-pauthtest %s 2>&1 | FileCheck %s --check-prefix=PAUTHABI3 +// PAUTHABI3: "-cc1" +// PAUTHABI3-SAME: "-target-abi" "aapcs" +// PAUTHABI3-NOT: "-fptrauth- +// PAUTHABI3-NOT: "-faarch64-jump-table-hardening" + // RUN: not %clang -### -c --target=x86_64 -fptrauth-intrinsics -fptrauth-calls -fptrauth-returns -fptrauth-auth-traps \ // RUN: -fptrauth-vtable-pointer-address-discrimination -fptrauth-vtable-pointer-type-discrimination \ // RUN: -fptrauth-type-info-vtable-pointer-discrimination -fptrauth-indirect-gotos -fptrauth-init-fini \ @@ -57,10 +71,12 @@ // ERR1-NEXT: error: unsupported option '-fptrauth-init-fini-address-discrimination' for target '{{.*}}' // ERR1-NEXT: error: unsupported option '-faarch64-jump-table-hardening' for target '{{.*}}' -//// Only support PAuth ABI for Linux as for now. -// RUN: not %clang -o /dev/null -c --target=aarch64-unknown -mabi=pauthtest %s 2>&1 | FileCheck %s --check-prefix=ERR2 -// RUN: not %clang -o /dev/null -c --target=aarch64-unknown-pauthtest %s 2>&1 | FileCheck %s --check-prefix=ERR2 -// ERR2: error: ABI 'pauthtest' is not supported for 'aarch64-unknown-unknown-pauthtest' + +// RUN: not %clang -c --target=aarch64 -mabi=pauthtest %s 2>&1 | FileCheck %s --check-prefix=ERR2 +//// The ABI is not specified explicitly, and for non-Linux pauthtest environment does not correspond +//// to pauthtest ABI (each OS target defines this behavior separately). Do not emit an error. +// RUN: %clang -c --target=aarch64-pauthtest %s -o /dev/null +// ERR2: error: unknown target ABI 'pauthtest' //// PAuth ABI is encoded as environment part of the triple, so don't allow to explicitly set other environments. // RUN: not %clang -### -c --target=aarch64-linux-gnu -mabi=pauthtest %s 2>&1 | FileCheck %s --check-prefix=ERR3 diff --git a/clang/test/Driver/fuchsia.c b/clang/test/Driver/fuchsia.c index 3fb2a94..d0fec18 100644 --- a/clang/test/Driver/fuchsia.c +++ b/clang/test/Driver/fuchsia.c @@ -312,3 +312,13 @@ // RUN: | FileCheck %s -check-prefix=CHECK-NOSTDLIB-NOLIBC // CHECK-NOSTDLIB-NOLIBC-NOT: "warning:" // CHECK-NOSTDLIB-NOLIBC-NOT: "error:" + +// RUN: not %clang -### %s --target=aarch64-unknown-fuchsia \ +// RUN: -fsanitize=safe-stack 2>&1 \ +// RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \ +// RUN: | FileCheck %s -check-prefix=CHECK-NONX86-SAFESTACK +// RUN: not %clang -### %s --target=riscv64-unknown-fuchsia \ +// RUN: -fsanitize=safe-stack 2>&1 \ +// RUN: -resource-dir=%S/Inputs/resource_dir_with_per_target_subdir \ +// RUN: | FileCheck %s -check-prefix=CHECK-NONX86-SAFESTACK +// CHECK-NONX86-SAFESTACK: error: unsupported option '-fsanitize=safe-stack' for target '{{.*}}' diff --git a/clang/test/Driver/hexagon-toolchain-elf.c b/clang/test/Driver/hexagon-toolchain-elf.c index de2ebfe..bff4819 100644 --- a/clang/test/Driver/hexagon-toolchain-elf.c +++ b/clang/test/Driver/hexagon-toolchain-elf.c @@ -166,6 +166,13 @@ // CHECK250: "-cc1" {{.*}} "-target-cpu" "hexagonv79" // CHECK250: hexagon-link{{.*}}/Inputs/hexagon_tree/Tools/bin/../target/hexagon/lib/v79/crt0 +// RUN: not %clang -### --target=hexagon-unknown-elf \ +// RUN: -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \ +// RUN: -mcpu=hexagonv81 -fuse-ld=hexagon-link \ +// RUN: %s 2>&1 | FileCheck -check-prefix=CHECK260 %s +// CHECK260: "-cc1" {{.*}} "-target-cpu" "hexagonv81" +// CHECK260: hexagon-link{{.*}}/Inputs/hexagon_tree/Tools/bin/../target/hexagon/lib/v81/crt0 + // ----------------------------------------------------------------------------- // Test Linker related args // ----------------------------------------------------------------------------- diff --git a/clang/test/Frontend/cfi-unchecked-callee-attribute.cpp b/clang/test/Frontend/cfi-unchecked-callee-attribute.cpp index 072f217..a5a17dd 100644 --- a/clang/test/Frontend/cfi-unchecked-callee-attribute.cpp +++ b/clang/test/Frontend/cfi-unchecked-callee-attribute.cpp @@ -9,6 +9,7 @@ void (*checked_ptr)(void) = unchecked; // expected-warning{{implicit conversion void (CFI_UNCHECKED_CALLEE *unchecked_ptr)(void) = unchecked; void (CFI_UNCHECKED_CALLEE *from_normal)(void) = checked; void (CFI_UNCHECKED_CALLEE *c_no_function_decay)(void) = &unchecked; +void (CFI_UNCHECKED_CALLEE __attribute__((noreturn)) *other_conflict)(void) = &checked; // expected-error{{cannot initialize a variable of type 'void (*)() __attribute__((noreturn)) __attribute__((cfi_unchecked_callee))' with an rvalue of type 'void (*)()'}} void (CFI_UNCHECKED_CALLEE *arr[10])(void); void (*cfi_elem)(void) = arr[1]; // expected-warning{{implicit conversion from 'void (*)() __attribute__((cfi_unchecked_callee))' to 'void (*)()' discards 'cfi_unchecked_callee' attribute}} void (CFI_UNCHECKED_CALLEE *cfi_unchecked_elem)(void) = arr[1]; diff --git a/clang/test/Preprocessor/hexagon-predefines.c b/clang/test/Preprocessor/hexagon-predefines.c index eebf481..bdedcf4 100644 --- a/clang/test/Preprocessor/hexagon-predefines.c +++ b/clang/test/Preprocessor/hexagon-predefines.c @@ -171,6 +171,23 @@ // CHECK-V79HVX-128B: #define __HVX__ 1 // CHECK-V79HVX-128B: #define __hexagon__ 1 +// RUN: %clang_cc1 -E -dM -triple hexagon-unknown-elf -target-cpu hexagonv81 %s\ +// RUN: | FileCheck %s -check-prefix CHECK-V81 +// CHECK-V81: #define __HEXAGON_ARCH__ 81 +// CHECK-V81: #define __HEXAGON_PHYSICAL_SLOTS__ 4 +// CHECK-V81: #define __HEXAGON_V81__ 1 +// CHECK-V81: #define __hexagon__ 1 + +// RUN: %clang_cc1 -E -dM -triple hexagon-unknown-elf -target-cpu hexagonv81 \ +// RUN: -target-feature +hvxv81 -target-feature +hvx-length128b %s | FileCheck \ +// RUN: %s -check-prefix CHECK-V81HVX-128B +// CHECK-V81HVX-128B: #define __HEXAGON_ARCH__ 81 +// CHECK-V81HVX-128B: #define __HEXAGON_V81__ 1 +// CHECK-V81HVX-128B: #define __HVX_ARCH__ 81 +// CHECK-V81HVX-128B: #define __HVX_LENGTH__ 128 +// CHECK-V81HVX-128B: #define __HVX__ 1 +// CHECK-V81HVX-128B: #define __hexagon__ 1 + // RUN: %clang_cc1 -E -dM -triple hexagon-unknown-elf -target-cpu hexagonv67 \ // RUN: -target-feature +hvxv67 -target-feature +hvx-length128b %s | FileCheck \ // RUN: %s -check-prefix CHECK-ELF diff --git a/clang/test/Sema/sifive-xsfmm.c b/clang/test/Sema/sifive-xsfmm.c new file mode 100644 index 0000000..7e055dd --- /dev/null +++ b/clang/test/Sema/sifive-xsfmm.c @@ -0,0 +1,30 @@ +// RUN: %clang_cc1 -triple riscv64 -target-feature +v \ +// RUN: -target-feature +xsfmmbase -target-feature +xsfmm32a -target-feature +xsfmm32a8f \ +// RUN: -target-feature +xsfmm32a16f -target-feature +xsfmm32a32f -target-feature +xsfmm64a64f \ +// RUN: -target-feature +xsfmm32a4f -target-feature +xsfmm32a8i -disable-O0-optnone \ +// RUN: -fsyntax-only %s -verify +// REQUIRES: riscv-registered-target +#include <sifive_vector.h> + +void test(vfloat32m8_t arg0, vuint8m8_t arg1) { + __riscv_sf_mm_f_f_w1(4, arg0, arg0, 1, 2, 3); + __riscv_sf_mm_e5m2_e4m3(8, arg1, arg1, 1, 2, 3); + __riscv_sf_mm_u_u(12, arg1, arg1, 1, 2, 3); + __riscv_sf_vtzero_t_e8w1(0, 0, 0); + + __riscv_sf_mm_f_f_w1(5, arg0, arg0, 1, 2, 3); /* expected-error {{argument should be a multiple of 4}} */ + __riscv_sf_mm_e5m2_e4m3(7, arg1, arg1, 1, 2, 3); /* expected-error {{argument should be a multiple of 4}} */ + __riscv_sf_mm_u_u(15, arg1, arg1, 1, 2, 3); /* expected-error {{argument should be a multiple of 4}} */ + __riscv_sf_mm_f_f_w1(16, arg0, arg0, 1, 2, 3); /* expected-error {{argument value 16 is outside the valid range [0, 15]}} */ + __riscv_sf_mm_e5m2_e4m3(20, arg1, arg1, 1, 2, 3); /* expected-error {{argument value 20 is outside the valid range [0, 15]}} */ + __riscv_sf_mm_u_u(24, arg1, arg1, 1, 2, 3); /* expected-error {{argument value 24 is outside the valid range [0, 15]}} */ + __riscv_sf_vtzero_t_e8w1(18, 0, 0); /* expected-error {{argument value 18 is outside the valid range [0, 15]}} */ + __riscv_sf_vtzero_t_e16w1(3, 0, 0); /* expected-error {{argument should be a multiple of 2}} */ + __riscv_sf_vtzero_t_e16w2(3, 0, 0); /* expected-error {{argument should be a multiple of 4}} */ + __riscv_sf_vtzero_t_e32w1(5, 0, 0); /* expected-error {{argument should be a multiple of 4}} */ + __riscv_sf_vtzero_t_e32w2(5, 0, 0); /* expected-error {{argument should be a multiple of 2}} */ + __riscv_sf_vtzero_t(5, 0, 0, 7, 1); /* expected-error {{argument value 7 is outside the valid range [3, 6]}} */ + __riscv_sf_vtzero_t(5, 0, 0, 2, 1); /* expected-error {{argument value 2 is outside the valid range [3, 6]}} */ + __riscv_sf_vtzero_t(5, 0, 0, 6, 3); /* expected-error {{RISC-V XSfmm twiden must be 1, 2 or 4}} */ + __riscv_sf_vtzero_t(5, 0, 0, 6, 5); /* expected-error {{RISC-V XSfmm twiden must be 1, 2 or 4}} */ +} diff --git a/clang/test/Sema/sifive_sf_vset_invalid.c b/clang/test/Sema/sifive_sf_vset_invalid.c new file mode 100644 index 0000000..96d8e0d --- /dev/null +++ b/clang/test/Sema/sifive_sf_vset_invalid.c @@ -0,0 +1,17 @@ +// REQUIRES: riscv-registered-target +// RUN: %clang_cc1 -triple riscv64 -target-feature +v \ +// RUN: -target-feature +xsfmmbase -disable-O0-optnone \ +// RUN: -o - -fsyntax-only %s -verify + +#include <sifive_vector.h> + +void test(size_t vl) { + __riscv_sf_vsettnt(vl, 1, 8); + // expected-error@-1 {{argument value 8 is outside the valid range [1, 3]}} + __riscv_sf_vsettm(vl, 8, 9); + // expected-error@-1 {{argument value 8 is outside the valid range [0, 3]}} + __riscv_sf_vsettn(vl, 8, 2); + // expected-error@-1 {{argument value 8 is outside the valid range [0, 3]}} + __riscv_sf_vsettk(vl, 0, 0); + // expected-error@-1 {{argument value 0 is outside the valid range [1, 3]}} +} diff --git a/clang/test/SemaHLSL/BuiltIns/matrix-constructors-errors.hlsl b/clang/test/SemaHLSL/BuiltIns/matrix-constructors-errors.hlsl new file mode 100644 index 0000000..e60a865 --- /dev/null +++ b/clang/test/SemaHLSL/BuiltIns/matrix-constructors-errors.hlsl @@ -0,0 +1,31 @@ +// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.0-compute -x hlsl -fsyntax-only -verify %s + +typedef float float2x1 __attribute__((matrix_type(2,1))); +typedef float float2x2 __attribute__((matrix_type(2,2))); +typedef float float2 __attribute__((ext_vector_type(2))); + +struct S { float f; }; +struct S2 { float2 f;}; + +[numthreads(1,1,1)] +void entry() { + float2x1 LilMat = float2x1(1.0, 2.0); + float2x1 BrokenMat = float2x1(1.0, 2.0, 3.0); // expected-error{{too many initializers in list for type 'float2x1' (aka 'matrix<float, 2, 1>') (expected 2 but found 3)}} + float2x2 NormieMat = float2x2(LilMat, 3.0, 4.0, 5.0); // expected-error{{too many initializers in list for type 'float2x2' (aka 'matrix<float, 2, 2>') (expected 4 but found 5)}} + float2x2 BrokenNormie = float2x2(3.0, 4.0); // expected-error{{too few initializers in list for type 'float2x2' (aka 'matrix<float, 2, 2>') (expected 4 but found 2)}} + float2x1 OverwhemledNormie = float2x1(3.0, 4.0, 5.0, 6.0); // expected-error{{too many initializers in list for type 'float2x1' (aka 'matrix<float, 2, 1>') (expected 2 but found 4)}} + + // These should work in HLSL and not error + S s; + float2x1 GettingStrange = float2x1(s, s); + + S2 s2; + float2x2 GettingStrange2 = float2x2(s2, s2); + + // HLSL does not yet allow user-defined conversions. + struct T { + operator float() const { return 1.0f; } + } t; + // TODO: Should this work? Today HLSL doesn't resolve user-defined conversions here, but we maybe should... + float2x1 foo5 = float2x1(t, t); // expected-error{{too few initializers in list for type 'float2x1' (aka 'matrix<float, 2, 1>') (expected 2 but found 0)}} +} diff --git a/clang/tools/driver/cc1_main.cpp b/clang/tools/driver/cc1_main.cpp index 49f8843..52cffa4 100644 --- a/clang/tools/driver/cc1_main.cpp +++ b/clang/tools/driver/cc1_main.cpp @@ -313,17 +313,6 @@ int cc1_main(ArrayRef<const char *> Argv, const char *Argv0, void *MainAddr) { llvm::TimerGroup::clearAll(); if (llvm::timeTraceProfilerEnabled()) { - // It is possible that the compiler instance doesn't own a file manager here - // if we're compiling a module unit. Since the file manager are owned by AST - // when we're compiling a module unit. So the file manager may be invalid - // here. - // - // It should be fine to create file manager here since the file system - // options are stored in the compiler invocation and we can recreate the VFS - // from the compiler invocation. - if (!Clang->hasFileManager()) - Clang->createFileManager(); - if (auto profilerOutput = Clang->createOutputFile( Clang->getFrontendOpts().TimeTracePath, /*Binary=*/false, /*RemoveFileOnSignal=*/false, diff --git a/clang/utils/TableGen/RISCVVEmitter.cpp b/clang/utils/TableGen/RISCVVEmitter.cpp index 74f29ac..c316dfd3 100644 --- a/clang/utils/TableGen/RISCVVEmitter.cpp +++ b/clang/utils/TableGen/RISCVVEmitter.cpp @@ -241,6 +241,8 @@ static unsigned getSegInstLog2SEW(StringRef InstName) { void emitCodeGenSwitchBody(const RVVIntrinsic *RVVI, raw_ostream &OS) { if (!RVVI->getIRName().empty()) OS << " ID = Intrinsic::riscv_" + RVVI->getIRName() + ";\n"; + if (RVVI->getTWiden() > 0) + OS << " TWiden = " << RVVI->getTWiden() << ";\n"; OS << " PolicyAttrs = " << RVVI->getPolicyAttrsBits() << ";\n"; unsigned IndexedLoadStorePtrIdx = getIndexedLoadStorePtrIdx(RVVI); @@ -295,6 +297,9 @@ void emitCodeGenSwitchBody(const RVVIntrinsic *RVVI, raw_ostream &OS) { OS << " Ops.insert(Ops.begin(), llvm::PoisonValue::get(ResultType));\n"; } + if (RVVI->getTWiden() > 0) + OS << " Ops.push_back(ConstantInt::get(Ops.back()->getType(), TWiden));\n"; + OS << " IntrinsicTypes = {"; ListSeparator LS; for (const auto &Idx : RVVI->getIntrinsicTypes()) { @@ -583,7 +588,8 @@ void RVVEmitter::createCodeGen(raw_ostream &OS) { (Def->getManualCodegen() != PrevDef->getManualCodegen()) || (Def->getPolicyAttrs() != PrevDef->getPolicyAttrs()) || (getSegInstLog2SEW(Def->getOverloadedName()) != - getSegInstLog2SEW(PrevDef->getOverloadedName()))) { + getSegInstLog2SEW(PrevDef->getOverloadedName())) || + (Def->getTWiden() != PrevDef->getTWiden())) { emitCodeGenSwitchBody(PrevDef, OS); } PrevDef = Def.get(); @@ -645,6 +651,7 @@ void RVVEmitter::createRVVIntrinsics( StringRef IRName = R->getValueAsString("IRName"); StringRef MaskedIRName = R->getValueAsString("MaskedIRName"); unsigned NF = R->getValueAsInt("NF"); + unsigned TWiden = R->getValueAsInt("TWiden"); bool IsTuple = R->getValueAsBit("IsTuple"); bool HasFRMRoundModeOp = R->getValueAsBit("HasFRMRoundModeOp"); @@ -694,7 +701,7 @@ void RVVEmitter::createRVVIntrinsics( /*IsMasked=*/false, /*HasMaskedOffOperand=*/false, HasVL, UnMaskedPolicyScheme, SupportOverloading, HasBuiltinAlias, ManualCodegen, *Types, IntrinsicTypes, NF, DefaultPolicy, - HasFRMRoundModeOp)); + HasFRMRoundModeOp, TWiden)); if (UnMaskedPolicyScheme != PolicyScheme::SchemeNone) for (auto P : SupportedUnMaskedPolicies) { SmallVector<PrototypeDescriptor> PolicyPrototype = @@ -709,7 +716,7 @@ void RVVEmitter::createRVVIntrinsics( /*IsMask=*/false, /*HasMaskedOffOperand=*/false, HasVL, UnMaskedPolicyScheme, SupportOverloading, HasBuiltinAlias, ManualCodegen, *PolicyTypes, IntrinsicTypes, NF, P, - HasFRMRoundModeOp)); + HasFRMRoundModeOp, TWiden)); } if (!HasMasked) continue; @@ -720,7 +727,7 @@ void RVVEmitter::createRVVIntrinsics( Name, SuffixStr, OverloadedName, OverloadedSuffixStr, MaskedIRName, /*IsMasked=*/true, HasMaskedOffOperand, HasVL, MaskedPolicyScheme, SupportOverloading, HasBuiltinAlias, ManualCodegen, *MaskTypes, - IntrinsicTypes, NF, DefaultPolicy, HasFRMRoundModeOp)); + IntrinsicTypes, NF, DefaultPolicy, HasFRMRoundModeOp, TWiden)); if (MaskedPolicyScheme == PolicyScheme::SchemeNone) continue; for (auto P : SupportedMaskedPolicies) { @@ -735,7 +742,7 @@ void RVVEmitter::createRVVIntrinsics( MaskedIRName, /*IsMasked=*/true, HasMaskedOffOperand, HasVL, MaskedPolicyScheme, SupportOverloading, HasBuiltinAlias, ManualCodegen, *PolicyTypes, IntrinsicTypes, NF, P, - HasFRMRoundModeOp)); + HasFRMRoundModeOp, TWiden)); } } // End for Log2LMULList } // End for TypeRange |