[libc] Add proxy header math_macros.h. #87598

lntue · 2024-04-04T05:24:57Z

Context: #87017

Add proxy header libc/hdr/math_macros.h that will:
- include <math.h> in overlay mode,
- include "include/llvm-libc-macros/math-macros.h" in full build mode.
Its corresponding CMake target libc.hdr.math_macros will only depend on libc.include.math and libc.include.llvm-libc-macros.math_macros in full build mode.
Replace all #include "include/llvm-libc-macros/math-macros.h" with #include "hdr/math_macros.h".
Add dependency to libc.hdr.math_macros CMake target when using add_fp_unittest.
Update the remaining dependency.
Update bazel overlay: add libc:hdr_math_macros target, and replacing all dependency on libc:llvm_libc_macros_math_macros with libc:hdr_math_macros.

llvmbot · 2024-04-04T05:25:35Z

@llvm/pr-subscribers-libc

Author: None (lntue)

Changes

Context: #87017

Add proxy header libc/hdr/math_macros.h that will:
- include <math.h> in overlay mode,
- include "include/llvm-libc-macros/math-macros.h" in full build mode.
Its corresponding CMake target libc.hdr.math_macros will only depend on libc.include.math and libc.include.llvm-libc-macros.math_macros in full build mode.
Replace all #include "include/llvm-libc-macros/math-macros.h" with #include "hdr/math_macros.h".
Add dependency to libc.hdr.math_macros CMake target when using add_fp_unittest.
Update the remaining dependency.
Update bazel overlay: add libc:hdr_math_macros target, and replacing all dependency on libc:llvm_libc_macros_math_macros with libc:hdr_math_macros.

Patch is 117.25 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/87598.diff

143 Files Affected:

(modified) libc/CMakeLists.txt (+1)
(modified) libc/fuzzing/math/CMakeLists.txt (+1)
(modified) libc/fuzzing/math/RemQuoDiff.h (+1-1)
(modified) libc/fuzzing/stdlib/CMakeLists.txt (+1)
(modified) libc/fuzzing/stdlib/strtofloat_fuzz.cpp (+1-1)
(added) libc/hdr/CMakeLists.txt (+33)
(added) libc/hdr/math_macros.h (+22)
(modified) libc/include/llvm-libc-macros/math-macros.h (+4-4)
(modified) libc/src/__support/FPUtil/CMakeLists.txt (+3-4)
(modified) libc/src/__support/FPUtil/FEnvImpl.h (+1-1)
(modified) libc/src/__support/FPUtil/ManipulationFunctions.h (+1-1)
(modified) libc/src/__support/FPUtil/NearestIntegerOperations.h (+1-1)
(modified) libc/src/math/generic/CMakeLists.txt (+1-16)
(modified) libc/src/math/generic/math_utils.h (+1-1)
(modified) libc/test/CMakeLists.txt (+2-4)
(modified) libc/test/UnitTest/FPMatcher.h (+1-1)
(modified) libc/test/src/CMakeLists.txt (+7-1)
(modified) libc/test/src/__support/uint_test.cpp (+1-1)
(modified) libc/test/src/math/CMakeLists.txt (-88)
(modified) libc/test/src/math/CeilTest.h (+1-1)
(modified) libc/test/src/math/CopySignTest.h (+1-1)
(modified) libc/test/src/math/FAbsTest.h (+1-1)
(modified) libc/test/src/math/FDimTest.h (+1-1)
(modified) libc/test/src/math/FMaxTest.h (+1-1)
(modified) libc/test/src/math/FMinTest.h (+1-1)
(modified) libc/test/src/math/FModTest.h (+1-1)
(modified) libc/test/src/math/FloorTest.h (+1-1)
(modified) libc/test/src/math/FrexpTest.h (+1-1)
(modified) libc/test/src/math/HypotTest.h (+1-1)
(modified) libc/test/src/math/ILogbTest.h (+1-1)
(modified) libc/test/src/math/LdExpTest.h (+1-1)
(modified) libc/test/src/math/LogbTest.h (+1-1)
(modified) libc/test/src/math/ModfTest.h (+1-1)
(modified) libc/test/src/math/NextAfterTest.h (+1-1)
(modified) libc/test/src/math/RIntTest.h (+1-1)
(modified) libc/test/src/math/RemQuoTest.h (+1-1)
(modified) libc/test/src/math/RoundTest.h (+1-1)
(modified) libc/test/src/math/RoundToIntegerTest.h (+1-1)
(modified) libc/test/src/math/SqrtTest.h (+1-1)
(modified) libc/test/src/math/TruncTest.h (+1-1)
(modified) libc/test/src/math/acosf_test.cpp (+1-1)
(modified) libc/test/src/math/acoshf_test.cpp (+1-1)
(modified) libc/test/src/math/asinf_test.cpp (+1-1)
(modified) libc/test/src/math/asinhf_test.cpp (+1-1)
(modified) libc/test/src/math/atan2f_test.cpp (+1-1)
(modified) libc/test/src/math/atanf_test.cpp (+1-1)
(modified) libc/test/src/math/atanhf_test.cpp (+1-1)
(modified) libc/test/src/math/cos_test.cpp (+1-1)
(modified) libc/test/src/math/cosf_test.cpp (+1-1)
(modified) libc/test/src/math/coshf_test.cpp (+1-1)
(modified) libc/test/src/math/erff_test.cpp (+1-1)
(modified) libc/test/src/math/exhaustive/CMakeLists.txt (-24)
(modified) libc/test/src/math/exp10_test.cpp (+1-1)
(modified) libc/test/src/math/exp10f_test.cpp (+1-1)
(modified) libc/test/src/math/exp2_test.cpp (+1-1)
(modified) libc/test/src/math/exp2f_test.cpp (+1-1)
(modified) libc/test/src/math/exp_test.cpp (+1-1)
(modified) libc/test/src/math/expf_test.cpp (+1-1)
(modified) libc/test/src/math/explogxf_test.cpp (+1-1)
(modified) libc/test/src/math/expm1_test.cpp (+1-1)
(modified) libc/test/src/math/expm1f_test.cpp (+1-1)
(modified) libc/test/src/math/fdim_test.cpp (+1-1)
(modified) libc/test/src/math/fdimf_test.cpp (+1-1)
(modified) libc/test/src/math/fdiml_test.cpp (+1-1)
(modified) libc/test/src/math/generic/CMakeLists.txt (-3)
(modified) libc/test/src/math/ilogb_test.cpp (+1-1)
(modified) libc/test/src/math/ilogbf_test.cpp (+1-1)
(modified) libc/test/src/math/ilogbl_test.cpp (+1-1)
(modified) libc/test/src/math/log10_test.cpp (+1-1)
(modified) libc/test/src/math/log10f_test.cpp (+1-1)
(modified) libc/test/src/math/log1p_test.cpp (+1-1)
(modified) libc/test/src/math/log1pf_test.cpp (+1-1)
(modified) libc/test/src/math/log2_test.cpp (+1-1)
(modified) libc/test/src/math/log2f_test.cpp (+1-1)
(modified) libc/test/src/math/log_test.cpp (+1-1)
(modified) libc/test/src/math/logf_test.cpp (+1-1)
(modified) libc/test/src/math/powf_test.cpp (+1-1)
(modified) libc/test/src/math/sin_test.cpp (+1-1)
(modified) libc/test/src/math/sincosf_test.cpp (+1-1)
(modified) libc/test/src/math/sinf_test.cpp (+1-1)
(modified) libc/test/src/math/sinhf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/CMakeLists.txt (-102)
(modified) libc/test/src/math/smoke/CanonicalizeTest.h (+1-1)
(modified) libc/test/src/math/smoke/CeilTest.h (+1-1)
(modified) libc/test/src/math/smoke/CopySignTest.h (+1-1)
(modified) libc/test/src/math/smoke/FAbsTest.h (+1-1)
(modified) libc/test/src/math/smoke/FModTest.h (+1-1)
(modified) libc/test/src/math/smoke/FloorTest.h (+1-1)
(modified) libc/test/src/math/smoke/HypotTest.h (+1-1)
(modified) libc/test/src/math/smoke/ModfTest.h (+1-1)
(modified) libc/test/src/math/smoke/NextAfterTest.h (+1-1)
(modified) libc/test/src/math/smoke/NextTowardTest.h (+1-1)
(modified) libc/test/src/math/smoke/RIntTest.h (+1-1)
(modified) libc/test/src/math/smoke/RemQuoTest.h (+1-1)
(modified) libc/test/src/math/smoke/RoundTest.h (+1-1)
(modified) libc/test/src/math/smoke/RoundToIntegerTest.h (+1-1)
(modified) libc/test/src/math/smoke/SqrtTest.h (+1-1)
(modified) libc/test/src/math/smoke/TruncTest.h (+1-1)
(modified) libc/test/src/math/smoke/acosf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/acoshf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/asinf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/asinhf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/atan2f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/atanf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/atanhf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/cosf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/coshf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/erff_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/exp10_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/exp10f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/exp2_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/exp2f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/exp_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/expf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/expm1_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/expm1f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log10_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log10f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log1p_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log1pf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log2_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log2f_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/log_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/logf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/powf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/sincosf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/sinf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/sinhf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/tanf_test.cpp (+1-1)
(modified) libc/test/src/math/smoke/tanhf_test.cpp (+1-1)
(modified) libc/test/src/math/tan_test.cpp (+1-1)
(modified) libc/test/src/math/tanf_test.cpp (+1-1)
(modified) libc/test/src/math/tanhf_test.cpp (+1-1)
(modified) libc/test/src/sys/random/linux/CMakeLists.txt (-1)
(modified) libc/test/utils/FPUtil/CMakeLists.txt (+1-1)
(modified) libc/test/utils/FPUtil/x86_long_double_test.cpp (+1-1)
(modified) libc/utils/MPFRWrapper/MPFRUtils.cpp (+1-1)
(modified) utils/bazel/llvm-project-overlay/libc/BUILD.bazel (+12-5)
(modified) utils/bazel/llvm-project-overlay/libc/test/UnitTest/BUILD.bazel (+1-1)
(modified) utils/bazel/llvm-project-overlay/libc/test/src/__support/BUILD.bazel (+1-1)
(modified) utils/bazel/llvm-project-overlay/libc/test/src/math/BUILD.bazel (+7-7)
(modified) utils/bazel/llvm-project-overlay/libc/test/src/math/libc_math_test_rules.bzl (+1-1)
(modified) utils/bazel/llvm-project-overlay/libc/utils/MPFRWrapper/BUILD.bazel (+1-1)

diff --git a/libc/CMakeLists.txt b/libc/CMakeLists.txt
index a0d79858a896ad..175efd89d67e6d 100644
--- a/libc/CMakeLists.txt
+++ b/libc/CMakeLists.txt
@@ -381,6 +381,7 @@ endforeach()
 
 add_subdirectory(include)
 add_subdirectory(config)
+add_subdirectory(hdr)
 add_subdirectory(src)
 add_subdirectory(utils)
 
diff --git a/libc/fuzzing/math/CMakeLists.txt b/libc/fuzzing/math/CMakeLists.txt
index 86c864083d206b..6990a04922a5cf 100644
--- a/libc/fuzzing/math/CMakeLists.txt
+++ b/libc/fuzzing/math/CMakeLists.txt
@@ -8,6 +8,7 @@ add_libc_fuzzer(
     SingleInputSingleOutputDiff.h
     TwoInputSingleOutputDiff.h
   DEPENDS
+    libc.hdr.math_macros
     libc.src.math.ceil
     libc.src.math.ceilf
     libc.src.math.ceill
diff --git a/libc/fuzzing/math/RemQuoDiff.h b/libc/fuzzing/math/RemQuoDiff.h
index 95a9866f29dbbe..84a6a24ce52719 100644
--- a/libc/fuzzing/math/RemQuoDiff.h
+++ b/libc/fuzzing/math/RemQuoDiff.h
@@ -11,7 +11,7 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include <stddef.h>
 #include <stdint.h>
 
diff --git a/libc/fuzzing/stdlib/CMakeLists.txt b/libc/fuzzing/stdlib/CMakeLists.txt
index 711b0fd9820f93..204bc619318da2 100644
--- a/libc/fuzzing/stdlib/CMakeLists.txt
+++ b/libc/fuzzing/stdlib/CMakeLists.txt
@@ -22,6 +22,7 @@ add_libc_fuzzer(
   SRCS
     strtofloat_fuzz.cpp
   DEPENDS
+    libc.hdr.math_macros
     libc.src.stdlib.atof
     libc.src.stdlib.strtof
     libc.src.stdlib.strtod
diff --git a/libc/fuzzing/stdlib/strtofloat_fuzz.cpp b/libc/fuzzing/stdlib/strtofloat_fuzz.cpp
index b000321854d16d..c158162ba6238e 100644
--- a/libc/fuzzing/stdlib/strtofloat_fuzz.cpp
+++ b/libc/fuzzing/stdlib/strtofloat_fuzz.cpp
@@ -16,7 +16,7 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include <stddef.h>
 #include <stdint.h>
 
diff --git a/libc/hdr/CMakeLists.txt b/libc/hdr/CMakeLists.txt
new file mode 100644
index 00000000000000..1d3eb9d5240ab9
--- /dev/null
+++ b/libc/hdr/CMakeLists.txt
@@ -0,0 +1,33 @@
+function(add_proxy_header_library target_name)
+  cmake_parse_arguments(
+    "ADD_PROXY_HEADER"
+    "" # Optional arguments
+    "" # Single value arguments
+    "DEPENDS;FULL_BUILD_DEPENDS" # Multi-value arguments
+    ${ARGN}
+  )
+
+  set(deps "")
+  if(ADD_PROXY_HEADER_DEPENDS)
+    list(APPEND deps ${ADD_PROXY_HEADER_DEPENDS})
+  endif()
+
+  if(LLVM_LIBC_FULL_BUILD AND ADD_PROXY_HEADER_FULL_BUILD_DEPENDS)
+    list(APPEND deps ${ADD_PROXY_HEADER_FULL_BUILD_DEPENDS})
+  endif()
+
+  add_header_library(
+    ${target_name}
+    ${ADD_PROXY_HEADER_UNPARSED_ARGUMENTS}
+    DEPENDS ${deps}
+  )
+endfunction()
+
+add_proxy_header_library(
+  math_macros
+  HDRS
+    math_macros.h
+  FULL_BUILD_DEPENDS
+    libc.include.llvm-libc-macros.math_macros
+    libc.include.math
+)
diff --git a/libc/hdr/math_macros.h b/libc/hdr/math_macros.h
new file mode 100644
index 00000000000000..008d00354b4b0e
--- /dev/null
+++ b/libc/hdr/math_macros.h
@@ -0,0 +1,22 @@
+//===-- Definition of macros from math.h ----------------------------------===//
+//
+// 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_HDR_MATH_MACROS_H
+#define LLVM_LIBC_HDR_MATH_MACROS_H
+
+#ifdef LLVM_LIBC_FULL_BUILD
+
+#include "include/llvm-libc-macros/math-macros.h"
+
+#else // Overlay mode
+
+#include <math.h>
+
+#endif // LLVM_LIBC_FULL_BUILD
+
+#endif // LLVM_LIBC_HDR_MATH_MACROS_H
diff --git a/libc/include/llvm-libc-macros/math-macros.h b/libc/include/llvm-libc-macros/math-macros.h
index 1497e32044e975..3ee32e331ed27c 100644
--- a/libc/include/llvm-libc-macros/math-macros.h
+++ b/libc/include/llvm-libc-macros/math-macros.h
@@ -30,12 +30,12 @@
 #define INFINITY __builtin_inf()
 #define NAN __builtin_nanf("")
 
-#define FP_ILOGB0 (-INT_MAX - 1)
-#define FP_LLOGB0 (-LONG_MAX - 1)
+#define FP_ILOGB0 INT_MIN
+#define FP_LLOGB0 LONG_MIN
 
 #ifdef __FP_LOGBNAN_MIN
-#define FP_ILOGBNAN (-INT_MAX - 1)
-#define FP_LLOGBNAN (-LONG_MAX - 1)
+#define FP_ILOGBNAN INT_MIN
+#define FP_LLOGBNAN LONG_MIN
 #else
 #define FP_ILOGBNAN INT_MAX
 #define FP_LLOGBNAN LONG_MAX
diff --git a/libc/src/__support/FPUtil/CMakeLists.txt b/libc/src/__support/FPUtil/CMakeLists.txt
index ff155a19758d20..1af236ed9d6bc1 100644
--- a/libc/src/__support/FPUtil/CMakeLists.txt
+++ b/libc/src/__support/FPUtil/CMakeLists.txt
@@ -4,7 +4,7 @@ add_header_library(
     FEnvImpl.h
   DEPENDS
     libc.include.fenv
-    libc.include.math
+    libc.hdr.math_macros
     libc.src.__support.macros.attributes
     libc.src.errno.errno
 )
@@ -15,7 +15,6 @@ add_header_library(
     rounding_mode.h
   DEPENDS
     libc.include.fenv
-    libc.include.math
     libc.src.__support.macros.attributes
     libc.src.__support.macros.properties.architectures
     libc.src.__support.macros.sanitizer
@@ -59,9 +58,9 @@ add_header_library(
     .fp_bits
     .fenv_impl
     .rounding_mode
+    libc.hdr.math_macros
     libc.src.__support.CPP.type_traits
     libc.src.__support.common
-    libc.include.math
     libc.src.errno.errno
 )
 
@@ -216,12 +215,12 @@ add_header_library(
     .dyadic_float
     .nearest_integer_operations
     .normal_float
+    libc.hdr.math_macros
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.limits
     libc.src.__support.CPP.type_traits
     libc.src.__support.common
     libc.src.__support.macros.optimization
-    libc.include.math
     libc.src.errno.errno
 )
 
diff --git a/libc/src/__support/FPUtil/FEnvImpl.h b/libc/src/__support/FPUtil/FEnvImpl.h
index 6086d5d3de2dca..4be1a57f0f4b38 100644
--- a/libc/src/__support/FPUtil/FEnvImpl.h
+++ b/libc/src/__support/FPUtil/FEnvImpl.h
@@ -9,7 +9,7 @@
 #ifndef LLVM_LIBC_SRC___SUPPORT_FPUTIL_FENVIMPL_H
 #define LLVM_LIBC_SRC___SUPPORT_FPUTIL_FENVIMPL_H
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include "src/__support/macros/attributes.h" // LIBC_INLINE
 #include "src/__support/macros/properties/architectures.h"
 #include "src/errno/libc_errno.h"
diff --git a/libc/src/__support/FPUtil/ManipulationFunctions.h b/libc/src/__support/FPUtil/ManipulationFunctions.h
index 2c90b4888c2e5d..a289c2ef70467b 100644
--- a/libc/src/__support/FPUtil/ManipulationFunctions.h
+++ b/libc/src/__support/FPUtil/ManipulationFunctions.h
@@ -15,7 +15,7 @@
 #include "dyadic_float.h"
 #include "rounding_mode.h"
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include "src/__support/CPP/bit.h"
 #include "src/__support/CPP/limits.h" // INT_MAX, INT_MIN
 #include "src/__support/CPP/type_traits.h"
diff --git a/libc/src/__support/FPUtil/NearestIntegerOperations.h b/libc/src/__support/FPUtil/NearestIntegerOperations.h
index 6b28e7ffb387b5..4645ab0b5350be 100644
--- a/libc/src/__support/FPUtil/NearestIntegerOperations.h
+++ b/libc/src/__support/FPUtil/NearestIntegerOperations.h
@@ -13,7 +13,7 @@
 #include "FPBits.h"
 #include "rounding_mode.h"
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include "src/__support/CPP/type_traits.h"
 #include "src/__support/common.h"
 
diff --git a/libc/src/math/generic/CMakeLists.txt b/libc/src/math/generic/CMakeLists.txt
index b164d33e204b1a..02760237d4f7b3 100644
--- a/libc/src/math/generic/CMakeLists.txt
+++ b/libc/src/math/generic/CMakeLists.txt
@@ -102,8 +102,8 @@ add_object_library(
   HDRS
     math_utils.h
   DEPENDS
+    libc.hdr.math_macros
     libc.include.errno
-    libc.include.math
     libc.src.errno.errno
 )
 
@@ -139,7 +139,6 @@ add_entrypoint_object(
     ../cosf.h
   DEPENDS
     .sincosf_utils
-    libc.include.math
     libc.src.errno.errno
     libc.src.__support.FPUtil.basic_operations
     libc.src.__support.FPUtil.fenv_impl
@@ -162,7 +161,6 @@ add_entrypoint_object(
   DEPENDS
     .range_reduction
     .sincosf_utils
-    libc.include.math
     libc.src.errno.errno
     libc.src.__support.FPUtil.basic_operations
     libc.src.__support.FPUtil.fenv_impl
@@ -185,7 +183,6 @@ add_entrypoint_object(
   DEPENDS
     .range_reduction
     .sincosf_utils
-    libc.include.math
     libc.src.errno.errno
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -207,7 +204,6 @@ add_entrypoint_object(
   DEPENDS
     .range_reduction
     .sincosf_utils
-    libc.include.math
     libc.src.errno.errno
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fenv_impl
@@ -721,7 +717,6 @@ add_entrypoint_object(
     libc.src.__support.FPUtil.multiply_add
     libc.src.__support.FPUtil.polyeval
     libc.src.__support.macros.optimization
-    libc.include.math
   COMPILE_OPTIONS
     -O3
 )
@@ -736,7 +731,6 @@ add_entrypoint_object(
     .common_constants
     .explogxf
     libc.include.errno
-    libc.include.math
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.optional
     libc.src.__support.FPUtil.dyadic_float
@@ -772,7 +766,6 @@ add_entrypoint_object(
     libc.src.__support.macros.optimization
     libc.include.errno
     libc.src.errno.errno
-    libc.include.math
   COMPILE_OPTIONS
     -O3
 )
@@ -787,7 +780,6 @@ add_entrypoint_object(
     .common_constants
     .explogxf
     libc.include.errno
-    libc.include.math
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.optional
     libc.src.__support.FPUtil.dyadic_float
@@ -822,7 +814,6 @@ add_header_library(
     libc.src.__support.common
     libc.include.errno
     libc.src.errno.errno
-    libc.include.math
 )
 
 add_entrypoint_object(
@@ -847,7 +838,6 @@ add_entrypoint_object(
     .common_constants
     .explogxf
     libc.include.errno
-    libc.include.math
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.optional
     libc.src.__support.FPUtil.dyadic_float
@@ -881,7 +871,6 @@ add_header_library(
     libc.src.__support.common
     libc.include.errno
     libc.src.errno.errno
-    libc.include.math
   COMPILE_OPTIONS
     -O3
 )
@@ -908,7 +897,6 @@ add_entrypoint_object(
     .common_constants
     .explogxf
     libc.include.errno
-    libc.include.math
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.optional
     libc.src.__support.FPUtil.dyadic_float
@@ -944,7 +932,6 @@ add_entrypoint_object(
     libc.src.__support.macros.optimization
     libc.include.errno
     libc.src.errno.errno
-    libc.include.math
   COMPILE_OPTIONS
     -O3
 )
@@ -961,7 +948,6 @@ add_entrypoint_object(
     .exp2f_impl
     .explogxf
     libc.include.errno
-    libc.include.math
     libc.src.__support.CPP.bit
     libc.src.__support.CPP.optional
     libc.src.__support.FPUtil.fenv_impl
@@ -2685,7 +2671,6 @@ add_object_library(
     libc.src.__support.common
     libc.include.errno
     libc.src.errno.errno
-    libc.include.math
   COMPILE_OPTIONS
     -O3
 )
diff --git a/libc/src/math/generic/math_utils.h b/libc/src/math/generic/math_utils.h
index cced761fc8c822..3ddfeccfd64837 100644
--- a/libc/src/math/generic/math_utils.h
+++ b/libc/src/math/generic/math_utils.h
@@ -9,7 +9,7 @@
 #ifndef LLVM_LIBC_SRC_MATH_GENERIC_MATH_UTILS_H
 #define LLVM_LIBC_SRC_MATH_GENERIC_MATH_UTILS_H
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 #include "src/__support/CPP/bit.h"
 #include "src/__support/CPP/type_traits.h"
 #include "src/__support/common.h"
diff --git a/libc/test/CMakeLists.txt b/libc/test/CMakeLists.txt
index 745a9a04b4af82..5e26a1000633b7 100644
--- a/libc/test/CMakeLists.txt
+++ b/libc/test/CMakeLists.txt
@@ -18,10 +18,6 @@ add_subdirectory(include)
 add_subdirectory(src)
 add_subdirectory(utils)
 
-if(LLVM_LIBC_FULL_BUILD AND NOT LIBC_TARGET_OS_IS_BAREMETAL)
-  add_subdirectory(IntegrationTest)
-endif()
-
 if(NOT LLVM_LIBC_FULL_BUILD)
   return()
 endif()
@@ -31,4 +27,6 @@ if(NOT ${LIBC_TARGET_OS} STREQUAL "linux" AND
   # Integration tests are currently only available for linux and the GPU.
   return()
 endif()
+
+add_subdirectory(IntegrationTest)
 add_subdirectory(integration)
diff --git a/libc/test/UnitTest/FPMatcher.h b/libc/test/UnitTest/FPMatcher.h
index f4553eac5c8a57..a76e0b8ef6f6f0 100644
--- a/libc/test/UnitTest/FPMatcher.h
+++ b/libc/test/UnitTest/FPMatcher.h
@@ -18,7 +18,7 @@
 #include "test/UnitTest/StringUtils.h"
 #include "test/UnitTest/Test.h"
 
-#include "include/llvm-libc-macros/math-macros.h"
+#include "hdr/math_macros.h"
 
 namespace LIBC_NAMESPACE {
 namespace testing {
diff --git a/libc/test/src/CMakeLists.txt b/libc/test/src/CMakeLists.txt
index f70ffda3f700e5..95f1768c96aac3 100644
--- a/libc/test/src/CMakeLists.txt
+++ b/libc/test/src/CMakeLists.txt
@@ -3,7 +3,7 @@ function(add_fp_unittest name)
     "MATH_UNITTEST"
     "NEED_MPFR;UNIT_TEST_ONLY;HERMETIC_TEST_ONLY" # Optional arguments
     "" # Single value arguments
-    "LINK_LIBRARIES" # Multi-value arguments
+    "LINK_LIBRARIES;DEPENDS" # Multi-value arguments
     ${ARGN}
   )
 
@@ -28,11 +28,17 @@ function(add_fp_unittest name)
   endif()
   list(APPEND MATH_UNITTEST_LINK_LIBRARIES LibcFPTestHelpers)
 
+  set(deps libc.hdr.math_macros)
+  if(MATH_UNITTEST_DEPENDS)
+    list(APPEND deps ${MATH_UNITTEST_DEPENDS})
+  endif()
+
   add_libc_test(
     ${name}
     ${test_type}
     LINK_LIBRARIES "${MATH_UNITTEST_LINK_LIBRARIES}"
     "${MATH_UNITTEST_UNPARSED_ARGUMENTS}"
+    DEPENDS "${deps}"
   )
 endfunction(add_fp_unittest)
 
diff --git a/libc/test/src/__support/uint_test.cpp b/libc/test/src/__support/uint_test.cpp
index 5764324ca28815..de56428e6bb4b0 100644
--- a/libc/test/src/__support/uint_test.cpp
+++ b/libc/test/src/__support/uint_test.cpp
@@ -10,7 +10,7 @@
 #include "src/__support/UInt.h"
 #include "src/__support/macros/properties/types.h" // LIBC_TYPES_HAS_INT128
 
-#include "include/llvm-libc-macros/math-macros.h" // HUGE_VALF, HUGE_VALF
+#include "hdr/math_macros.h" // HUGE_VALF, HUGE_VALF
 #include "test/UnitTest/Test.h"
 
 namespace LIBC_NAMESPACE {
diff --git a/libc/test/src/math/CMakeLists.txt b/libc/test/src/math/CMakeLists.txt
index f8f0f8ba7b6f63..fa3e7f58309050 100644
--- a/libc/test/src/math/CMakeLists.txt
+++ b/libc/test/src/math/CMakeLists.txt
@@ -108,7 +108,6 @@ add_fp_unittest(
   HDRS
     FAbsTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.fabs
     libc.src.__support.FPUtil.fp_bits
 )
@@ -123,7 +122,6 @@ add_fp_unittest(
   HDRS
     FAbsTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.fabsf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -138,7 +136,6 @@ add_fp_unittest(
   HDRS
     FAbsTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.fabsl
     libc.src.__support.FPUtil.fp_bits
 )
@@ -153,7 +150,6 @@ add_fp_unittest(
   HDRS
     TruncTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.trunc
     libc.src.__support.FPUtil.fp_bits
 )
@@ -168,7 +164,6 @@ add_fp_unittest(
   HDRS
     TruncTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.truncf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -183,7 +178,6 @@ add_fp_unittest(
   HDRS
     TruncTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.truncl
     libc.src.__support.FPUtil.fp_bits
 )
@@ -198,7 +192,6 @@ add_fp_unittest(
   HDRS
     CeilTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.ceil
     libc.src.__support.FPUtil.fp_bits
 )
@@ -213,7 +206,6 @@ add_fp_unittest(
   HDRS
     CeilTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.ceilf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -228,7 +220,6 @@ add_fp_unittest(
   HDRS
     CeilTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.ceill
     libc.src.__support.FPUtil.fp_bits
 )
@@ -243,7 +234,6 @@ add_fp_unittest(
   HDRS
     FloorTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.floor
     libc.src.__support.FPUtil.fp_bits
 )
@@ -258,7 +248,6 @@ add_fp_unittest(
   HDRS
     FloorTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.floorf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -273,7 +262,6 @@ add_fp_unittest(
   HDRS
     FloorTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.floorl
     libc.src.__support.FPUtil.fp_bits
 )
@@ -288,7 +276,6 @@ add_fp_unittest(
   HDRS
     RoundTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.round
     libc.src.__support.FPUtil.fp_bits
 )
@@ -303,7 +290,6 @@ add_fp_unittest(
   HDRS
     RoundTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.roundf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -318,7 +304,6 @@ add_fp_unittest(
   HDRS
     RoundTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.roundl
     libc.src.__support.FPUtil.fp_bits
 )
@@ -333,7 +318,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -352,7 +336,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -371,7 +354,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -390,7 +372,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -409,7 +390,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -428,7 +408,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.errno.errno
     libc.src.fenv.feclearexcept
     libc.src.fenv.feraiseexcept
@@ -447,7 +426,6 @@ add_fp_unittest(
   HDRS
     RIntTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.rint
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -463,7 +441,6 @@ add_fp_unittest(
   HDRS
     RIntTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.rintf
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -479,7 +456,6 @@ add_fp_unittest(
   HDRS
     RIntTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.rintl
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -495,7 +471,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.lrint
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -511,7 +486,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.lrintf
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -527,7 +501,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.lrintl
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -543,7 +516,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.llrint
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -559,7 +531,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.llrintf
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -575,7 +546,6 @@ add_fp_unittest(
   HDRS
     RoundToIntegerTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.llrintl
     libc.src.__support.FPUtil.fenv_impl
     libc.src.__support.FPUtil.fp_bits
@@ -590,7 +560,6 @@ add_fp_unittest(
     expf_test.cpp
   DEPENDS
     libc.src.errno.errno
-    libc.include.math
     libc.src.math.expf
     libc.src.__support.FPUtil.fp_bits
 )
@@ -618,7 +587,6 @@ add_fp_unittest(
     exp2f_test.cpp
   DEPENDS
     libc.src.errno.errno
-    libc.include.math
     libc.src.math.exp2f
     libc.src.__support.FPUtil.fp_bits
 )
@@ -646,7 +614,6 @@ add_fp_unittest(
     exp10f_test.cpp
   DEPENDS
     libc.src.errno.errno
-    libc.include.math
     libc.src.math.exp10f
     libc.src.__support.FPUtil.fp_bits
 )
@@ -674,7 +641,6 @@ add_fp_unittest(
   HDRS
     CopySignTest.h
   DEPENDS
-    libc.include.math
     libc.src.math.copysign...
[truncated]

SchrodingerZhu

The patch overall looks good to me. Do we expose any macro to end users to indicate full build and overlay build?

lntue · 2024-04-04T13:12:37Z

The patch overall looks good to me. Do we expose any macro to end users to indicate full build and overlay build?

For cmake cli, -DLLVM_LIBC_FULL_BUILD=ON should do it. For user codes:

Once everything is routed through proxy headers, there won't be any include/* headers, and users will just include system headers.
For full build mode, since our include/* headers will now only be included in this mode, and they can assume that we are the only headers of the system, and don't need to do any of the #include_next dances.

For implementations in src folder, we can use #ifdef LIBC_FULL_BUILD to check whether we are in full build mode or overlay mode.

SchrodingerZhu

LGTM

nickdesaulniers

Thanks! I think this gives me a clearer picture of what the usage of "proxy" headers might look like (#87017).

nickdesaulniers · 2024-04-04T16:27:04Z

libc/hdr/CMakeLists.txt

+    ${ADD_PROXY_HEADER_UNPARSED_ARGUMENTS}
+    DEPENDS ${deps}
+  )
+endfunction()


ah, nice, I should rebase #87017 onto this cmake change.

libc/include/llvm-libc-macros/math-macros.h

…its.

lntue requested review from nickdesaulniers and michaelrj-google April 4, 2024 05:24

lntue requested a review from rupprecht as a code owner April 4, 2024 05:24

llvmbot added libc bazel "Peripheral" support tier build system: utils/bazel labels Apr 4, 2024

lntue requested review from SchrodingerZhu, gchatelet and jhuber6 April 4, 2024 05:25

lntue requested review from petrhosek and frobtech April 4, 2024 05:25

SchrodingerZhu reviewed Apr 4, 2024

View reviewed changes

lntue force-pushed the math_macros branch from 911308c to 8be421e Compare April 4, 2024 12:43

SchrodingerZhu approved these changes Apr 4, 2024

View reviewed changes

nickdesaulniers mentioned this pull request Apr 4, 2024

[RFC][libc] Codify header inclusion policy #87017

Merged

nickdesaulniers reviewed Apr 4, 2024

View reviewed changes

lntue force-pushed the math_macros branch from ee50768 to 97f8e48 Compare April 4, 2024 17:35

This was referenced Apr 4, 2024

[libc] FP_ILOGBNAN / FP_LLOGBNAN compat #87683

Open

[libc] remove MPFR and related tests in full build #87693

Merged

[libc] Temporary math macros fix #87681

Merged

lntue force-pushed the math_macros branch from 97f8e48 to c54f2e7 Compare April 5, 2024 15:51

lntue added 7 commits April 5, 2024 20:55

[libc] Add proxy header math_macros.h.

4a20fda

Use LIBC_FULL_BUILD flag.

176e580

Sync and remove remaining libc.include.math in tests.

bfe02f6

Restore math-macros.h.

dbb57cf

Sync

7cd02e4

Sync and remove LIBC_FULL_BUILD check in math-macros.h.

c94d5c0

Update some "include/llvm-libc-macros/math-macros.h" from recent comm…

1a5b952

…its.

lntue force-pushed the math_macros branch from c54f2e7 to 1a5b952 Compare April 5, 2024 20:58

nickdesaulniers approved these changes Apr 5, 2024

View reviewed changes

michaelrj-google approved these changes Apr 5, 2024

View reviewed changes

lntue merged commit 5748ad8 into llvm:main Apr 5, 2024
4 checks passed

lntue deleted the math_macros branch April 5, 2024 23:27

This was referenced Apr 6, 2024

[libc] Add proxy header for fenv.h macro constants. #87863

Closed

[libc] Add proxy headers for fenv.h types. #88187

Closed

[libc] Add proxy header hdr/float-macros.h #88433

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[libc] Add proxy header math_macros.h. #87598

[libc] Add proxy header math_macros.h. #87598

lntue commented Apr 4, 2024

llvmbot commented Apr 4, 2024

SchrodingerZhu left a comment •

edited

Loading

lntue commented Apr 4, 2024

SchrodingerZhu left a comment

nickdesaulniers left a comment

nickdesaulniers Apr 4, 2024

[libc] Add proxy header math_macros.h. #87598

[libc] Add proxy header math_macros.h. #87598

Conversation

lntue commented Apr 4, 2024

llvmbot commented Apr 4, 2024

SchrodingerZhu left a comment • edited Loading

Choose a reason for hiding this comment

lntue commented Apr 4, 2024

SchrodingerZhu left a comment

Choose a reason for hiding this comment

nickdesaulniers left a comment

Choose a reason for hiding this comment

nickdesaulniers Apr 4, 2024

Choose a reason for hiding this comment

SchrodingerZhu left a comment •

edited

Loading