Add acosf16() function #127725
Closed
Add acosf16() function #127725
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
wldfngrs
commented
Feb 19, 2025
wldfngrs
commented
- Implementation of acosf16() function.
- Exhaustive tests across 16-bit input range.
|
@llvm/pr-subscribers-libc Author: wldfngrs (wldfngrs) Changes
Full diff: https://github.com/llvm/llvm-project/pull/127725.diff 12 Files Affected:
diff --git a/libc/config/linux/x86_64/entrypoints.txt b/libc/config/linux/x86_64/entrypoints.txt
index a4f6671a59789..b38fe7520e293 100644
--- a/libc/config/linux/x86_64/entrypoints.txt
+++ b/libc/config/linux/x86_64/entrypoints.txt
@@ -648,6 +648,7 @@ if(LIBC_TYPES_HAS_FLOAT16)
list(APPEND TARGET_LIBM_ENTRYPOINTS
# math.h C23 _Float16 entrypoints
libc.src.math.asinf16
+ libc.src.math.acosf16
libc.src.math.canonicalizef16
libc.src.math.ceilf16
libc.src.math.copysignf16
diff --git a/libc/docs/headers/math/index.rst b/libc/docs/headers/math/index.rst
index 3e45e3e618abb..3010f9457ae11 100644
--- a/libc/docs/headers/math/index.rst
+++ b/libc/docs/headers/math/index.rst
@@ -250,7 +250,7 @@ Higher Math Functions
+-----------+------------------+-----------------+------------------------+----------------------+------------------------+------------------------+----------------------------+
| <Func> | <Func_f> (float) | <Func> (double) | <Func_l> (long double) | <Func_f16> (float16) | <Func_f128> (float128) | C23 Definition Section | C23 Error Handling Section |
+===========+==================+=================+========================+======================+========================+========================+============================+
-| acos | |check| | | | | | 7.12.4.1 | F.10.1.1 |
+| acos | |check| | | | |check| | | 7.12.4.1 | F.10.1.1 |
+-----------+------------------+-----------------+------------------------+----------------------+------------------------+------------------------+----------------------------+
| acosh | |check| | | | | | 7.12.5.1 | F.10.2.1 |
+-----------+------------------+-----------------+------------------------+----------------------+------------------------+------------------------+----------------------------+
diff --git a/libc/include/math.yaml b/libc/include/math.yaml
index b98bc55f6cc53..a66f981030864 100644
--- a/libc/include/math.yaml
+++ b/libc/include/math.yaml
@@ -14,6 +14,13 @@ functions:
return_type: float
arguments:
- type: float
+ - name: acosf16
+ standards:
+ - stdc
+ return_type: _Float16
+ arguments:
+ - type: _Float16
+ guard: LIBC_TYPES_HAS_FLOAT16
- name: acoshf
standards:
- stdc
diff --git a/libc/src/math/CMakeLists.txt b/libc/src/math/CMakeLists.txt
index 82551a4b57f45..f18a73d46f9aa 100644
--- a/libc/src/math/CMakeLists.txt
+++ b/libc/src/math/CMakeLists.txt
@@ -42,6 +42,8 @@ endfunction()
add_math_entrypoint_object(acos)
add_math_entrypoint_object(acosf)
+add_math_entrypoint_object(acosf16)
+
add_math_entrypoint_object(acosh)
add_math_entrypoint_object(acoshf)
diff --git a/libc/src/math/acosf16.h b/libc/src/math/acosf16.h
new file mode 100644
index 0000000000000..df30be2c537c9
--- /dev/null
+++ b/libc/src/math/acosf16.h
@@ -0,0 +1,21 @@
+//===-- Implementation header for acosf16 -----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_SRC_MATH_ACOSF16_H
+#define LLVM_LIBC_SRC_MATH_ACOSF16_H
+
+#include "src/__support/macros/config.h"
+#include "src/__support/macros/properties/types.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+float16 acosf16(float16 x);
+
+} // namespace LIBC_NAMESPACE_DECL
+
+#endif // LLVM_LIBC_SRC_MATH_ACOSF16_H
diff --git a/libc/src/math/generic/CMakeLists.txt b/libc/src/math/generic/CMakeLists.txt
index 537d5b5ad94ed..079e03d953d1c 100644
--- a/libc/src/math/generic/CMakeLists.txt
+++ b/libc/src/math/generic/CMakeLists.txt
@@ -4034,6 +4034,26 @@ add_entrypoint_object(
.inv_trigf_utils
)
+add_entrypoint_object(
+ acosf16
+ SRCS
+ acosf16.cpp
+ HDRS
+ ../acosf16.h
+ DEPENDS
+ libc.hdr.errno_macros
+ libc.hdr.fenv_macros
+ libc.src.__support.FPUtil.cast
+ libc.src.__support.FPUtil.fenv_impl
+ libc.src.__support.FPUtil.fp_bits
+ libc.src.__support.FPUtil.except_value_utils
+ libc.src.__support.FPUtil.multiply_add
+ libc.src.__support.FPUtil.polyeval
+ libc.src.__support.FPUtil.sqrt
+ libc.src.__support.macros.optimization
+ libc.src.__support.macros.properties.types
+)
+
add_entrypoint_object(
atanf
SRCS
diff --git a/libc/src/math/generic/acosf16.cpp b/libc/src/math/generic/acosf16.cpp
new file mode 100644
index 0000000000000..94f6dfddce20d
--- /dev/null
+++ b/libc/src/math/generic/acosf16.cpp
@@ -0,0 +1,146 @@
+//===-- Half-precision acosf16(x) function --------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception.
+//
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/math/acosf16.h"
+#include "hdr/errno_macros.h"
+#include "hdr/fenv_macros.h"
+#include "src/__support/FPUtil/FEnvImpl.h"
+#include "src/__support/FPUtil/FPBits.h"
+#include "src/__support/FPUtil/PolyEval.h"
+#include "src/__support/FPUtil/cast.h"
+#include "src/__support/FPUtil/except_value_utils.h"
+#include "src/__support/FPUtil/multiply_add.h"
+#include "src/__support/FPUtil/sqrt.h"
+#include "src/__support/macros/optimization.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+// Generated by Sollya using the following command:
+// > round(pi/2, D, RN);
+// > round(pi, D, RN);
+static constexpr float PI_2 = 0x1.921fb54442d18p0f;
+static constexpr float PI = 0x1.921fb54442d18p1f;
+
+static constexpr size_t N_EXCEPTS = 2;
+
+static constexpr fputil::ExceptValues<float16, N_EXCEPTS> ACOSF16_EXCEPTS{{
+ // (input, RZ output, RU offset, RD offset, RN offset)
+ {0xacaf, 0x3e93, 1, 0, 0},
+ {0xb874, 0x4052, 1, 0, 1},
+}};
+
+LLVM_LIBC_FUNCTION(float16, acosf16, (float16 x)) {
+ using FPBits = fputil::FPBits<float16>;
+ FPBits xbits(x);
+
+ uint16_t x_u = xbits.uintval();
+ uint16_t x_abs = x_u & 0x7fff;
+ uint16_t x_sign = x_u >> 15;
+ float xf = x;
+
+ // Handle exceptional values
+ if (auto r = ACOSF16_EXCEPTS.lookup(x_u); LIBC_UNLIKELY(r.has_value()))
+ return r.value();
+
+ // |x| == 0x1p0, x is 1 or -1
+ // if x is (-)1, return pi, else
+ // if x is (+)1, return 0
+ if (LIBC_UNLIKELY(x_abs == 0x3c00))
+ return fputil::cast<float16>(x_sign ? 0x1.921fb54442d18p1 : 0.0f);
+
+ // |x| > 0x1p0, |x| > 1, or x is NaN.
+ if (LIBC_UNLIKELY(x_abs > 0x3c00)) {
+ // acosf16(NaN) = NaN
+ if (xbits.is_nan()) {
+ if (xbits.is_signaling_nan()) {
+ fputil::raise_except_if_required(FE_INVALID);
+ return FPBits::quiet_nan().get_val();
+ }
+
+ return x;
+ }
+
+ // 1 < |x| <= +/-inf
+ fputil::raise_except_if_required(FE_INVALID);
+ fputil::set_errno_if_required(EDOM);
+
+ return FPBits::quiet_nan().get_val();
+ }
+
+ float xsq = xf * xf;
+
+ // |x| <= 0x1p-1, |x| <= 0.5
+ if (x_abs <= 0x3800) {
+ // if x is 0, return pi/2
+ if (LIBC_UNLIKELY(x_abs == 0))
+ return fputil::cast<float16>(PI_2);
+
+ // Note that: acos(x) = pi/2 + asin(-x) = pi/2 - asin(x)
+ // Degree-6 minimax polynomial of asin(x) generated by Sollya with:
+ // > P = fpminimax(asin(x)/x, [|0, 2, 4, 6, 8|], [|SG...|], [0, 0.5]);
+ float interm =
+ fputil::polyeval(xsq, 0x1.000002p0f, 0x1.554c2ap-3f, 0x1.3541ccp-4f,
+ 0x1.43b2d6p-5f, 0x1.a0d73ep-5f);
+ return fputil::cast<float16>(fputil::multiply_add(-xf, interm, PI_2));
+ }
+
+ // When |x| > 0.5, assume that 0.5 < |x| <= 1
+ //
+ // Step-by-step range-reduction proof:
+ // 1: Let y = asin(x), such that, x = sin(y)
+ // 2: From complimentary angle identity:
+ // x = sin(y) = cos(pi/2 - y)
+ // 3: Let z = pi/2 - y, such that x = cos(z)
+ // 4: From double angle formula; cos(2A) = 1 - sin^2(A):
+ // z = 2A, z/2 = A
+ // cos(z) = 1 - 2 * sin*2(z/2)
+ // 5: Make sin(z/2) subject of the formula:
+ // sin(z/2) = sqrt((1 - cos(z))/2)
+ // 6: Recall [3]; x = cos(z). Therefore:
+ // sin(z/2) = sqrt((1 - x)/2)
+ // 7: Let u = (1 - x)/2
+ // 8: Therefore:
+ // asin(sqrt(u)) = z/2
+ // 2 * asin(sqrt(u)) = z
+ // 9: Recall [3]; z = pi/2 - y. Therefore:
+ // y = pi/2 - z
+ // y = pi/2 - 2 * asin(sqrt(u))
+ // 10: Recall [1], y = asin(x). Therefore:
+ // asin(x) = pi/2 - 2 * asin(sqrt(u))
+ // 11: Recall that: acos(x) = pi/2 + asin(-x) = pi/2 - asin(x)
+ // Therefore:
+ // acos(x) = pi/2 - (pi/2 - 2 * asin(sqrt(u)))
+ // acos(x) = 2 * asin(sqrt(u))
+ //
+ // THE RANGE REDUCTION, HOW?
+ // 12: Recall [7], u = (1 - x)/2
+ // 13: Since 0.5 < x <= 1, therefore:
+ // 0 <= u <= 0.25 and 0 <= sqrt(u) <= 0.5
+ //
+ // Hence, we can reuse the same [0, 0.5] domain polynomial approximation for
+ // Step [11] as `sqrt(u)` is in range.
+ // When -1 < x <= -0.5, the identity:
+ // acos(x) = pi - acos(-x)
+ // allows us to compute for the negative x value (lhs)
+ // with a positive x value instead (rhs).
+
+ float xf_abs = (xf < 0 ? -xf : xf);
+ float u = fputil::multiply_add(-0.5f, xf_abs, 0.5f);
+ float sqrt_u = fputil::sqrt<float>(u);
+
+ // Degree-6 minimax polynomial of asin(x) generated by Sollya with:
+ // > P = fpminimax(asin(x)/x, [|0, 2, 4, 6, 8|], [|SG...|], [0, 0.5]);
+ float asin_sqrt_u =
+ sqrt_u * fputil::polyeval(u, 0x1.000002p0f, 0x1.554c2ap-3f,
+ 0x1.3541ccp-4f, 0x1.43b2d6p-5f, 0x1.a0d73ep-5f);
+
+ return fputil::cast<float16>(
+ x_sign ? fputil::multiply_add(-2.0f, asin_sqrt_u, PI) : 2 * asin_sqrt_u);
+}
+} // namespace LIBC_NAMESPACE_DECL
diff --git a/libc/src/math/generic/tanf16.cpp b/libc/src/math/generic/tanf16.cpp
index 48aa51e456a8a..08eaee2d06f86 100644
--- a/libc/src/math/generic/tanf16.cpp
+++ b/libc/src/math/generic/tanf16.cpp
@@ -19,9 +19,9 @@
namespace LIBC_NAMESPACE_DECL {
-constexpr size_t N_EXCEPTS = 9;
+static constexpr size_t N_EXCEPTS = 9;
-constexpr fputil::ExceptValues<float16, N_EXCEPTS> TANF16_EXCEPTS{{
+static constexpr fputil::ExceptValues<float16, N_EXCEPTS> TANF16_EXCEPTS{{
// (input, RZ output, RU offset, RD offset, RN offset)
{0x2894, 0x2894, 1, 0, 1},
{0x3091, 0x3099, 1, 0, 0},
diff --git a/libc/test/src/math/CMakeLists.txt b/libc/test/src/math/CMakeLists.txt
index 6a3dd8c9deff0..53ddd301900c0 100644
--- a/libc/test/src/math/CMakeLists.txt
+++ b/libc/test/src/math/CMakeLists.txt
@@ -2210,6 +2210,17 @@ add_fp_unittest(
libc.src.__support.FPUtil.fp_bits
)
+add_fp_unittest(
+ acosf16_test
+ NEED_MPFR
+ SUITE
+ libc-math-unittests
+ SRCS
+ acosf16_test.cpp
+ DEPENDS
+ libc.src.math.acosf16
+)
+
add_fp_unittest(
atanf_test
NEED_MPFR
diff --git a/libc/test/src/math/acosf16_test.cpp b/libc/test/src/math/acosf16_test.cpp
new file mode 100644
index 0000000000000..9624cd6d01f1a
--- /dev/null
+++ b/libc/test/src/math/acosf16_test.cpp
@@ -0,0 +1,42 @@
+//===-- Exhaustive test for asinf16 ---------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/math/acosf16.h"
+#include "test/UnitTest/FPMatcher.h"
+#include "test/UnitTest/Test.h"
+#include "utils/MPFRWrapper/MPFRUtils.h"
+
+using LlvmLibcAcosf16Test = LIBC_NAMESPACE::testing::FPTest<float16>;
+
+namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
+
+// Range: [0, Inf]
+static constexpr uint16_t POS_START = 0x0000U;
+static constexpr uint16_t POS_STOP = 0x7c00U;
+
+// Range: [-Inf, 0]
+static constexpr uint16_t NEG_START = 0x8000U;
+static constexpr uint16_t NEG_STOP = 0xfc00U;
+
+TEST_F(LlvmLibcAcosf16Test, PositiveRange) {
+ for (uint16_t v = POS_START; v <= POS_STOP; ++v) {
+ float16 x = FPBits(v).get_val();
+
+ EXPECT_MPFR_MATCH_ALL_ROUNDING(mpfr::Operation::Acos, x,
+ LIBC_NAMESPACE::acosf16(x), 0.5);
+ }
+}
+
+TEST_F(LlvmLibcAcosf16Test, NegativeRange) {
+ for (uint16_t v = NEG_START; v <= NEG_STOP; ++v) {
+ float16 x = FPBits(v).get_val();
+
+ EXPECT_MPFR_MATCH_ALL_ROUNDING(mpfr::Operation::Acos, x,
+ LIBC_NAMESPACE::acosf16(x), 0.5);
+ }
+}
diff --git a/libc/test/src/math/smoke/CMakeLists.txt b/libc/test/src/math/smoke/CMakeLists.txt
index 14447728fb18a..6f94440d826d9 100644
--- a/libc/test/src/math/smoke/CMakeLists.txt
+++ b/libc/test/src/math/smoke/CMakeLists.txt
@@ -3980,6 +3980,17 @@ add_fp_unittest(
libc.src.__support.FPUtil.fp_bits
)
+add_fp_unittest(
+ acosf16_test
+ SUITE
+ libc-math-smoke-tests
+ SRCS
+ acosf16_test.cpp
+ DEPENDS
+ libc.src.errno.errno
+ libc.src.math.acosf16
+)
+
add_fp_unittest(
atanf_test
SUITE
diff --git a/libc/test/src/math/smoke/acosf16_test.cpp b/libc/test/src/math/smoke/acosf16_test.cpp
new file mode 100644
index 0000000000000..5518a0bace1d4
--- /dev/null
+++ b/libc/test/src/math/smoke/acosf16_test.cpp
@@ -0,0 +1,39 @@
+//===-- Unittests for acosf16 ---------------------------------------------===//
+//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception.
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/errno/libc_errno.h"
+#include "src/math/acosf16.h"
+#include "test/UnitTest/FPMatcher.h"
+#include "test/UnitTest/Test.h"
+
+using LlvmLibcAcosf16Test = LIBC_NAMESPACE::testing::FPTest<float16>;
+
+TEST_F(LlvmLibcAcosf16Test, SpecialNumbers) {
+ LIBC_NAMESPACE::libc_errno = 0;
+ EXPECT_FP_EQ(aNaN, LIBC_NAMESPACE::acosf16(aNaN));
+ EXPECT_MATH_ERRNO(0);
+
+ EXPECT_FP_EQ_WITH_EXCEPTION(aNaN, LIBC_NAMESPACE::acosf16(sNaN), FE_INVALID);
+ EXPECT_MATH_ERRNO(0);
+
+ EXPECT_FP_EQ(aNaN, LIBC_NAMESPACE::acosf16(inf));
+ EXPECT_MATH_ERRNO(EDOM);
+
+ EXPECT_FP_EQ(aNaN, LIBC_NAMESPACE::acosf16(neg_inf));
+ EXPECT_MATH_ERRNO(EDOM);
+
+ EXPECT_FP_EQ(zero, LIBC_NAMESPACE::acosf16(1.0f));
+ EXPECT_MATH_ERRNO(0);
+
+ EXPECT_FP_EQ(aNaN, LIBC_NAMESPACE::acosf16(2.0f));
+ EXPECT_MATH_ERRNO(EDOM);
+
+ EXPECT_FP_EQ(aNaN, LIBC_NAMESPACE::acosf16(-2.0f));
+ EXPECT_MATH_ERRNO(EDOM);
+}
|
You can test this locally with the following command:git-clang-format --diff ef9f0b3c414a5d55e694829514d7b2ff8736d3c3 effd133ee620059c6e38c28cfff8566bc1353b81 --extensions c,h,cpp -- clang/test/Preprocessor/riscv-cf-protection-return.c clang/test/Sema/gh106576.c clang/test/SemaCXX/coroutine-noreturn.cpp libc/src/math/acosf16.h libc/src/math/generic/acosf16.cpp libc/test/src/math/acosf16_test.cpp libc/test/src/math/smoke/acosf16_test.cpp lldb/unittests/Core/TelemetryTest.cpp llvm/include/llvm/Transforms/Vectorize/SandboxVectorizer/Passes/SeedCollection.h llvm/include/llvm/Transforms/Vectorize/SandboxVectorizer/Passes/TransactionSave.h llvm/lib/Target/AMDGPU/GCNNSAReassign.h llvm/lib/Transforms/Vectorize/SandboxVectorizer/Passes/SeedCollection.cpp llvm/lib/Transforms/Vectorize/SandboxVectorizer/Passes/TransactionSave.cpp bolt/tools/binary-analysis/binary-analysis.cpp bolt/tools/driver/llvm-bolt.cpp bolt/tools/heatmap/heatmap.cpp bolt/tools/llvm-bolt-fuzzer/llvm-bolt-fuzzer.cpp bolt/unittests/Core/BinaryContext.cpp bolt/unittests/Core/MCPlusBuilder.cpp bolt/unittests/Core/MemoryMaps.cpp clang-tools-extra/clangd/ClangdServer.h clang-tools-extra/clangd/Compiler.h clang/include/clang/AST/DeclBase.h clang/include/clang/AST/DeclCXX.h clang/include/clang/AST/Expr.h clang/include/clang/AST/ExprCXX.h clang/include/clang/AST/RecursiveASTVisitor.h clang/include/clang/Sema/Sema.h clang/include/clang/Sema/Template.h clang/include/clang/Serialization/ASTBitCodes.h clang/lib/AST/ASTContext.cpp clang/lib/AST/ByteCode/Compiler.cpp clang/lib/AST/ByteCode/Descriptor.cpp clang/lib/AST/ByteCode/Interp.cpp clang/lib/AST/ByteCode/Interp.h clang/lib/AST/DeclBase.cpp clang/lib/AST/DeclCXX.cpp clang/lib/AST/Expr.cpp clang/lib/AST/ExprCXX.cpp clang/lib/AST/ExprClassification.cpp clang/lib/AST/ExprConstant.cpp clang/lib/AST/ItaniumMangle.cpp clang/lib/AST/StmtPrinter.cpp clang/lib/AST/StmtProfile.cpp clang/lib/AST/TextNodeDumper.cpp clang/lib/Basic/Targets/RISCV.cpp clang/lib/CodeGen/CGBuiltin.cpp clang/lib/CodeGen/CGOpenMPRuntime.cpp clang/lib/Format/TokenAnnotator.cpp clang/lib/Frontend/CompilerInvocation.cpp clang/lib/Index/IndexBody.cpp clang/lib/Index/IndexingContext.cpp clang/lib/Index/IndexingContext.h clang/lib/Sema/AnalysisBasedWarnings.cpp clang/lib/Sema/SemaCoroutine.cpp clang/lib/Sema/SemaDeclCXX.cpp clang/lib/Sema/SemaExceptionSpec.cpp clang/lib/Sema/SemaExpr.cpp clang/lib/Sema/SemaStmt.cpp clang/lib/Sema/SemaTemplateInstantiate.cpp clang/lib/Sema/SemaTemplateInstantiateDecl.cpp clang/lib/Sema/SemaTemplateVariadic.cpp clang/lib/Sema/TreeTransform.h clang/lib/Serialization/ASTReaderStmt.cpp clang/lib/Serialization/ASTWriter.cpp clang/lib/Serialization/ASTWriterStmt.cpp clang/lib/StaticAnalyzer/Checkers/ArrayBoundChecker.cpp clang/lib/StaticAnalyzer/Core/ExprEngine.cpp clang/test/AST/ByteCode/records.cpp clang/test/AST/ast-dump-binding-pack.cpp clang/test/AST/ast-dump-decl.cpp clang/test/Analysis/out-of-bounds.c clang/test/Index/Core/index-dependent-source.cpp clang/test/SemaCXX/cxx2c-binding-pack.cpp clang/tools/clang-scan-deps/ClangScanDeps.cpp clang/tools/libclang/CXCursor.cpp clang/unittests/Format/FormatTest.cpp clang/unittests/Format/TokenAnnotatorTest.cpp flang-rt/include/flang-rt/runtime/io-stmt.h flang/include/flang/Lower/AbstractConverter.h flang/lib/Lower/Bridge.cpp flang/lib/Lower/OpenMP/ClauseProcessor.cpp flang/lib/Lower/OpenMP/ClauseProcessor.h flang/lib/Lower/OpenMP/OpenMP.cpp flang/lib/Lower/OpenMP/Utils.cpp flang/lib/Lower/OpenMP/Utils.h flang/lib/Optimizer/OpenMP/MapInfoFinalization.cpp flang/lib/Optimizer/OpenMP/MapsForPrivatizedSymbols.cpp libc/src/math/generic/tanf16.cpp libclc/clc/include/clc/clc_convert.h libcxx/include/__mbstate_t.h libcxx/src/experimental/tzdb.cpp libcxx/test/libcxx/gdb/gdb_pretty_printer_test.sh.cpp lldb/include/lldb/Breakpoint/BreakpointList.h lldb/include/lldb/Breakpoint/BreakpointLocationCollection.h lldb/include/lldb/Breakpoint/BreakpointLocationList.h lldb/include/lldb/Breakpoint/WatchpointList.h lldb/include/lldb/Breakpoint/WatchpointResource.h lldb/include/lldb/Core/ModuleList.h lldb/include/lldb/Core/ModuleSpec.h lldb/include/lldb/Core/Telemetry.h lldb/include/lldb/Host/common/NativeProcessProtocol.h lldb/include/lldb/Symbol/ObjectFile.h lldb/include/lldb/Symbol/SymbolContext.h lldb/include/lldb/Symbol/TypeList.h lldb/include/lldb/Symbol/TypeMap.h lldb/include/lldb/Target/QueueList.h lldb/include/lldb/Target/TargetList.h lldb/include/lldb/Target/ThreadCollection.h lldb/include/lldb/Utility/Iterable.h lldb/source/Core/Telemetry.cpp lldb/source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp lldb/source/Plugins/Process/mach-core/ProcessMachCore.cpp lldb/source/Plugins/Process/mach-core/ProcessMachCore.h llvm/include/llvm/Analysis/CallGraph.h llvm/include/llvm/Analysis/CallGraphSCCPass.h llvm/include/llvm/CodeGen/ISDOpcodes.h llvm/include/llvm/CodeGen/SelectionDAG.h llvm/include/llvm/CodeGen/TargetLowering.h llvm/include/llvm/CodeGen/TargetRegisterInfo.h llvm/include/llvm/Frontend/OpenMP/OMPIRBuilder.h llvm/include/llvm/InitializePasses.h llvm/include/llvm/Passes/CodeGenPassBuilder.h llvm/include/llvm/SandboxIR/Region.h llvm/include/llvm/TargetParser/RISCVTargetParser.h llvm/include/llvm/Transforms/Vectorize/SandboxVectorizer/Passes/BottomUpVec.h llvm/lib/Analysis/CallGraph.cpp llvm/lib/Analysis/CallGraphSCCPass.cpp llvm/lib/Analysis/InlineCost.cpp llvm/lib/Analysis/LoopAccessAnalysis.cpp llvm/lib/Analysis/ValueTracking.cpp llvm/lib/CodeGen/AsmPrinter/DbgEntityHistoryCalculator.cpp llvm/lib/CodeGen/MLRegAllocEvictAdvisor.cpp llvm/lib/CodeGen/RegAllocEvictionAdvisor.cpp llvm/lib/CodeGen/RegAllocGreedy.cpp llvm/lib/CodeGen/RegAllocGreedy.h llvm/lib/CodeGen/RegAllocPriorityAdvisor.h llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp llvm/lib/CodeGen/SelectionDAG/LegalizeTypes.h llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp llvm/lib/CodeGen/TargetLoweringBase.cpp llvm/lib/ExecutionEngine/Orc/ELFNixPlatform.cpp llvm/lib/ExecutionEngine/Orc/JITLinkRedirectableSymbolManager.cpp llvm/lib/ExecutionEngine/Orc/LazyReexports.cpp llvm/lib/Frontend/OpenMP/OMPIRBuilder.cpp llvm/lib/Passes/PassBuilder.cpp llvm/lib/ProfileData/ProfileSummaryBuilder.cpp llvm/lib/SandboxIR/Region.cpp llvm/lib/Target/AArch64/AArch64ISelLowering.cpp llvm/lib/Target/AMDGPU/AMDGPU.h llvm/lib/Target/AMDGPU/AMDGPULowerBufferFatPointers.cpp llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.cpp llvm/lib/Target/AMDGPU/AMDGPUMCInstLower.h llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h llvm/lib/Target/AMDGPU/AMDGPUWaitSGPRHazards.cpp llvm/lib/Target/AMDGPU/GCNNSAReassign.cpp llvm/lib/Target/AMDGPU/SIFoldOperands.cpp llvm/lib/Target/AMDGPU/SIInstrInfo.cpp llvm/lib/Target/AMDGPU/SIInstrInfo.h llvm/lib/Target/AMDGPU/SIProgramInfo.cpp llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h llvm/lib/Target/ARM/ARMISelLowering.cpp llvm/lib/Target/DirectX/DXILRootSignature.cpp llvm/lib/Target/Hexagon/HexagonISelDAGToDAG.cpp llvm/lib/Target/Hexagon/HexagonISelLowering.cpp llvm/lib/Target/Hexagon/HexagonVectorCombine.cpp llvm/lib/Target/NVPTX/MCTargetDesc/NVPTXInstPrinter.cpp llvm/lib/Target/PowerPC/MCTargetDesc/PPCInstPrinter.cpp llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCCodeEmitter.cpp llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.cpp llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.h llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp llvm/lib/Target/PowerPC/PPCFrameLowering.cpp llvm/lib/Target/PowerPC/PPCISelLowering.cpp llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp llvm/lib/Target/RISCV/MCA/RISCVCustomBehaviour.cpp llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h llvm/lib/Target/RISCV/MCTargetDesc/RISCVInstPrinter.cpp llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp llvm/lib/Target/RISCV/RISCVISelLowering.cpp llvm/lib/Target/RISCV/RISCVISelLowering.h llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp llvm/lib/Target/RISCV/RISCVInstrInfo.cpp llvm/lib/Target/RISCV/RISCVInstrInfo.h llvm/lib/Target/RISCV/RISCVRegisterInfo.h llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp llvm/lib/Target/RISCV/RISCVVectorMaskDAGMutation.cpp llvm/lib/Target/RISCV/RISCVVectorPeephole.cpp llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp llvm/lib/Target/VE/VEISelLowering.cpp llvm/lib/Target/X86/X86FastISel.cpp llvm/lib/Target/X86/X86FrameLowering.cpp llvm/lib/Target/X86/X86ISelLowering.cpp llvm/lib/Target/X86/X86InstrBuilder.h llvm/lib/Target/X86/X86InstrInfo.cpp llvm/lib/Target/X86/X86InstrInfo.h llvm/lib/Target/X86/X86LoadValueInjectionLoadHardening.cpp llvm/lib/TargetParser/RISCVTargetParser.cpp llvm/lib/Transforms/IPO/AlwaysInliner.cpp llvm/lib/Transforms/IPO/FunctionAttrs.cpp llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp llvm/lib/Transforms/Scalar/DeadStoreElimination.cpp llvm/lib/Transforms/Vectorize/LoopVectorize.cpp llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp llvm/lib/Transforms/Vectorize/SandboxVectorizer/DependencyGraph.cpp llvm/lib/Transforms/Vectorize/SandboxVectorizer/Passes/BottomUpVec.cpp llvm/lib/Transforms/Vectorize/SandboxVectorizer/SandboxVectorizer.cpp llvm/lib/Transforms/Vectorize/SandboxVectorizer/SandboxVectorizerPassBuilder.cpp llvm/lib/Transforms/Vectorize/VPlan.cpp llvm/lib/Transforms/Vectorize/VPlanHCFGBuilder.cpp llvm/lib/Transforms/Vectorize/VPlanHCFGBuilder.h llvm/unittests/Frontend/OpenMPIRBuilderTest.cpp llvm/unittests/SandboxIR/RegionTest.cpp llvm/unittests/TargetParser/RISCVTargetParserTest.cpp llvm/unittests/Transforms/Vectorize/SandboxVectorizer/DependencyGraphTest.cpp llvm/utils/TableGen/Common/CodeGenSchedule.h llvm/utils/TableGen/InstrInfoEmitter.cpp llvm/utils/TableGen/SubtargetEmitter.cpp mlir/include/mlir/Analysis/DataFlowFramework.h mlir/include/mlir/Dialect/EmitC/IR/EmitC.h mlir/include/mlir/Dialect/GPU/IR/CompilationInterfaces.h mlir/include/mlir/Dialect/Math/Transforms/Approximation.h mlir/include/mlir/Dialect/Math/Transforms/Passes.h mlir/include/mlir/Dialect/SCF/Utils/Utils.h mlir/include/mlir/Dialect/Tosa/Utils/ConversionUtils.h mlir/include/mlir/IR/OpImplementation.h mlir/include/mlir/Target/Cpp/CppEmitter.h mlir/include/mlir/Target/LLVMIR/ModuleImport.h mlir/lib/Conversion/MathToLibm/MathToLibm.cpp mlir/lib/Conversion/OpenMPToLLVM/OpenMPToLLVM.cpp mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp mlir/lib/Dialect/EmitC/IR/EmitC.cpp mlir/lib/Dialect/GPU/IR/GPUDialect.cpp mlir/lib/Dialect/Linalg/IR/LinalgInterfaces.cpp mlir/lib/Dialect/Math/IR/MathOps.cpp mlir/lib/Dialect/Math/Transforms/PolynomialApproximation.cpp mlir/lib/Dialect/OpenMP/IR/OpenMPDialect.cpp mlir/lib/Dialect/SCF/Utils/Utils.cpp mlir/lib/Dialect/SPIRV/IR/SPIRVDialect.cpp mlir/lib/Dialect/Tensor/IR/TensorOps.cpp mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp mlir/lib/Dialect/Tosa/IR/TosaOps.cpp mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp mlir/lib/Dialect/Tosa/Utils/ConversionUtils.cpp mlir/lib/Dialect/X86Vector/Transforms/LegalizeForLLVMExport.cpp mlir/lib/IR/AsmPrinter.cpp mlir/lib/Target/Cpp/TranslateRegistration.cpp mlir/lib/Target/Cpp/TranslateToCpp.cpp mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp mlir/lib/Target/LLVMIR/ModuleImport.cpp mlir/test/lib/Dialect/SCF/TestLoopUnrolling.cpp mlir/test/lib/Dialect/Test/TestAttributes.cpp mlir/test/lib/Dialect/Test/TestTypes.cpp offload/libomptarget/OpenMP/Mapping.cpp utils/bazel/llvm-project-overlay/llvm/include/llvm/Config/config.h llvm/include/llvm/CodeGen/RegAllocEvictionAdvisor.hView the diff from clang-format here.diff --git a/libc/src/math/generic/acosf16.cpp b/libc/src/math/generic/acosf16.cpp
index 94f6dfddce..4f9e9a2fb7 100644
--- a/libc/src/math/generic/acosf16.cpp
+++ b/libc/src/math/generic/acosf16.cpp
@@ -30,9 +30,9 @@ static constexpr float PI = 0x1.921fb54442d18p1f;
static constexpr size_t N_EXCEPTS = 2;
static constexpr fputil::ExceptValues<float16, N_EXCEPTS> ACOSF16_EXCEPTS{{
- // (input, RZ output, RU offset, RD offset, RN offset)
- {0xacaf, 0x3e93, 1, 0, 0},
- {0xb874, 0x4052, 1, 0, 1},
+ // (input, RZ output, RU offset, RD offset, RN offset)
+ {0xacaf, 0x3e93, 1, 0, 0},
+ {0xb874, 0x4052, 1, 0, 1},
}};
LLVM_LIBC_FUNCTION(float16, acosf16, (float16 x)) {
diff --git a/libcxx/src/experimental/tzdb.cpp b/libcxx/src/experimental/tzdb.cpp
index 1f18226636..9c82a468a5 100644
--- a/libcxx/src/experimental/tzdb.cpp
+++ b/libcxx/src/experimental/tzdb.cpp
@@ -764,8 +764,8 @@ void __init_tzdb(tzdb& __tzdb, __tz::__rules_storage_type& __rules) {
// zone.
//
// On Linux systems it seems /etc/timezone is deprecated and being phased
- // out. This file is used when /etc/localtime does not exist, or when it exists but is not a symlink. For more information and links see
- // https://github.com/llvm/llvm-project/issues/105634
+ // out. This file is used when /etc/localtime does not exist, or when it exists but is not a symlink. For more
+ // information and links see https://github.com/llvm/llvm-project/issues/105634
string __name = chrono::__current_zone_environment();
|
…5034) tests to be added while porting virtregrewrite and greedy regalloc
The uses replace hand-rolled code that did a subset of what HeuristicResolver does.
Introduces a `LinalgInterfaces` target so that `TensorDialect` doesn't need to depend on `LinalgDialect`, which would introduce a circular dependency.
… pseudos. (llvm#127535) ForceTailAgnostic is currently treated as an override of the policy operand. It doesn't do anything else so we can just use the policy directly.
And returns true for div/rem/sqrt/... operations. This is an alternative if we don't support generic scheduling model.
This include introduced an unwanted dependency from tensor to tensor utils.
llvm#121386 Introduced cttz intrinsics which caused a regression where vscale/vscale divisions could no longer be constant folded. This fold was suggested as a fix in llvm#126411. https://alive2.llvm.org/ce/z/gWbtPw
…127158) Some loops cannot be unrolled by affine-loop-unroll pass. After running lower-affine pass, they can be unrolled in scf.To enable conversion of vector Ops in scf to llvm dialect, unroll-full option was added. --------- Co-authored-by: Oleksandr "Alex" Zinenko <[email protected]>
…m#127185) Silence compiler warning introduced in llvm#125007 - assign the address delta to int64_t, assert it is negative and negate it only as part of the mergeSPAdd call Fixes llvm#125825
…ling (llvm#127530) Support extension to 512-bit vectors on AVX512/BWI targets.
Also ensure we can break before ClassHeadName like StartOfName. Fixes llvm#127470
Bolt currently links and initializes all LLVM targets. This substantially increases the binary size, and link time if LTO is used. Instead, only link the targets specified by BOLT_TARGETS_TO_BUILD. We also have to only initialize those targets, so generate a TargetConfig.def file with the necessary information. The way the initialization is done mirrors what llvm-exegesis does. This reduces llvm-bolt size from 137MB to 78MB for me.
Use getSignedConstant() in a few more places, based on a search of `\bgetConstant(-`. Most of these were fine as-is (e.g. because they work on 64-bits), but I think it's better to use getSignedConstant() consistently for negative numbers.
We don't really care of this is an inline constant, only if it will be legal.
…rmPackExpr (llvm#125394) This merges the functionality of ResolvedUnexpandedPackExpr into FunctionParmPackExpr. I also added a test to show that llvm#125103 should be fixed with this. I put the removal of ResolvedUnexpandedPackExpr in its own commit. Let me know what you think. Fixes llvm#125103
…ic type (llvm#119711) When a statement expression's last statement is an atomic variable, GCC and Clang disagree on the type of the expression. This can be made apparent using `typeof` and forcing a diagnostic message: ```cpp _Atomic int a = 0; typeof(({a;})) x = "0"; ``` * GCC complains about initializing `int` with `char*` * Clang complains about initializing `_Atomic(int)` with a `char[2]` Due to the type of the statement expression being deduced to be atomic, we end with three implicit casts inside the `StmtExpr` on the AST: * `LValueToRValue` -> `AtomicToNonAtomic` -> `NonAtomicToAtomic` In some situations, this can end on an assertion inside `IntExprEvaluator`, as reported in llvm#106576. With this patch, we now have two implicit casts, since the type of the statement expression is deduced to be non-atomic: * `LValueToRValue` -> `AtomicToNonAtomic` This is consistent with the C standard (6.7.2.4, p4) > The properties associated with atomic types are meaningful only for expressions that are lvalues. But a statement expression is an rvalue. `IntExprEvaluator` assumptions are now satisfied and there is no assertion error. Additionally, the `typeof` trick mentioned above shows that the type is consistently deduced between GCC and Clang. Fixes llvm#106576 --------- Co-authored-by: John McCall <[email protected]>
…m#125901) This should further reduce the number of spurious barriers
Add signed and unsigned PARTIAL_REDUCE_MLA ISD nodes. Add command line argument (aarch64-enable-partial-reduce-nodes) that indicates whether the intrinsic experimental_vector_partial_ reduce_add will be transformed into the new ISD node. Lowering with the new ISD nodes will, for now, always be done as an expand.
…ray" (llvm#127596) Reverts llvm#125901 Revert until I have fixed bot failures
These targets don't include all OpenCL builtins, so there will always be external calls in the final bytecode module. Fixes llvm#127316.
…m#127587) expsensive_check complains that: bb.0.entry: %0:gpr32 = RDHWR $hwr2, 0 %1:gpr32 = ADDiu $zero, 0 $v0 = COPY %0:gpr32 $v1 = COPY %1:gpr32 RetRA implicit $v0, implicit $v1 *** Bad machine code: Using an undefined physical register *** - function: test_readcyclecounter - basic block: %bb.0 entry (0xad97ee0) - instruction: %0:gpr32 = RDHWR $hwr2, 0 - operand 1: $hwr2 LLVM ERROR: Found 1 machine code errors.
The OpenCL C specification states that for out-of-range dimension indices, `get_num_groups` must return 1 instead of 0.
… stack is enabled (llvm#127592) The `-fcf-protection=[full|return]` flag enables shadow stack implementation based on RISC-V Zicfiss extension. This patch adds the `__riscv_shadow_stack` predefined macro to preprocessing when such a shadow stack implementation is enabled.
The docs conflate `__funcref` with an actual type in a couple places.
…ion" (llvm#127673) Previous patch is merged llvm#114500 and it hit a buildbot failure and thus reverted It seems the AMDGPU::OpName::OPERAND_LAST is removed at the meantime when previous patch is merged and that's causing the compile error. Fixed and reopen it here
…124956) Move the `scale`, `offset`, and `border` parameters of the RESIZE operator in the MLIR TOSA dialect from attributes to inputs and update lit tests appropriately. Add the verifier of the `tosa::ResizeOp` operation. --------- Co-authored-by: Tai Ly <[email protected]> Co-authored-by: Luke Hutton <[email protected]>
wldfngrs
requested review from
rupprecht,
keith,
aaronmondal,
aaupov,
maksfb,
rafaelauler,
ayermolo,
dcci,
yota9,
antiagainst,
kuhar,
aartbik,
dcaballe,
nicolasvasilache,
rengolin,
JDevlieghere,
nikic and
a team
as code owners
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.