[libc][math][c23] Add {,u}fromfp{,x}{,f,l,f128} functions #86003
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
|
@llvm/pr-subscribers-libc Author: OverMighty (overmighty) ChangesFixes #85279. cc @lntue Patch is 143.85 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/86003.diff 63 Files Affected:
diff --git a/libc/config/linux/x86_64/entrypoints.txt b/libc/config/linux/x86_64/entrypoints.txt
index e8cf11266624a7..d17240ab8de521 100644
--- a/libc/config/linux/x86_64/entrypoints.txt
+++ b/libc/config/linux/x86_64/entrypoints.txt
@@ -383,6 +383,12 @@ set(TARGET_LIBM_ENTRYPOINTS
libc.src.math.frexp
libc.src.math.frexpf
libc.src.math.frexpl
+ libc.src.math.fromfp
+ libc.src.math.fromfpf
+ libc.src.math.fromfpl
+ libc.src.math.fromfpx
+ libc.src.math.fromfpxf
+ libc.src.math.fromfpxl
libc.src.math.hypot
libc.src.math.hypotf
libc.src.math.ilogb
@@ -467,6 +473,12 @@ set(TARGET_LIBM_ENTRYPOINTS
libc.src.math.trunc
libc.src.math.truncf
libc.src.math.truncl
+ libc.src.math.ufromfp
+ libc.src.math.ufromfpf
+ libc.src.math.ufromfpl
+ libc.src.math.ufromfpx
+ libc.src.math.ufromfpxf
+ libc.src.math.ufromfpxl
)
if(LIBC_TYPES_HAS_FLOAT128)
@@ -481,6 +493,8 @@ if(LIBC_TYPES_HAS_FLOAT128)
libc.src.math.fminf128
libc.src.math.fmodf128
libc.src.math.frexpf128
+ libc.src.math.fromfpf128
+ libc.src.math.fromfpxf128
libc.src.math.ilogbf128
libc.src.math.ldexpf128
libc.src.math.llogbf128
@@ -498,6 +512,8 @@ if(LIBC_TYPES_HAS_FLOAT128)
libc.src.math.roundf128
libc.src.math.sqrtf128
libc.src.math.truncf128
+ libc.src.math.ufromfpf128
+ libc.src.math.ufromfpxf128
)
endif()
diff --git a/libc/docs/math/index.rst b/libc/docs/math/index.rst
index d337d060fb5dd9..fa604faaec2ff0 100644
--- a/libc/docs/math/index.rst
+++ b/libc/docs/math/index.rst
@@ -181,6 +181,22 @@ Basic Operations
+--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
| frexpf128 | |check| | |check| | | |check| | | | | | | | | |
+--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfp | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpf | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpl | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpf128 | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpx | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpxf | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpxl | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| fromfpxf128 | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
| ilogb | |check| | |check| | |check| | |check| | |check| | | | |check| | |check| | |check| | | |
+--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
| ilogbf | |check| | |check| | |check| | |check| | |check| | | | |check| | |check| | |check| | | |
@@ -339,6 +355,22 @@ Basic Operations
+--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
| truncf128 | |check| | |check| | | |check| | | | | | | | | |
+--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfp | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpf | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpl | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpf128 | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpx | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpxf | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpxl | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
+| ufromfpxf128 | |check| | | | | | | | | | | | |
++--------------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+
Higher Math Functions
diff --git a/libc/include/llvm-libc-macros/math-macros.h b/libc/include/llvm-libc-macros/math-macros.h
index db8a4ea65bd69a..0d89f020ebb8e6 100644
--- a/libc/include/llvm-libc-macros/math-macros.h
+++ b/libc/include/llvm-libc-macros/math-macros.h
@@ -17,6 +17,12 @@
#define FP_SUBNORMAL 3
#define FP_NORMAL 4
+#define FP_INT_UPWARD 0
+#define FP_INT_DOWNWARD 1
+#define FP_INT_TOWARDZERO 2
+#define FP_INT_TONEARESTFROMZERO 3
+#define FP_INT_TONEAREST 4
+
#define MATH_ERRNO 1
#define MATH_ERREXCEPT 2
diff --git a/libc/spec/stdc.td b/libc/spec/stdc.td
index 84d28cc3350304..5f425ba1120a23 100644
--- a/libc/spec/stdc.td
+++ b/libc/spec/stdc.td
@@ -352,6 +352,12 @@ def StdC : StandardSpec<"stdc"> {
Macro<"INFINITY">,
Macro<"NAN">,
+ Macro<"FP_INT_UPWARD">,
+ Macro<"FP_INT_DOWNWARD">,
+ Macro<"FP_INT_TOWARDZERO">,
+ Macro<"FP_INT_TONEARESTFROMZERO">,
+ Macro<"FP_INT_TONEAREST">,
+
Macro<"FP_ILOGB0">,
Macro<"FP_ILOGBNAN">,
@@ -414,6 +420,26 @@ def StdC : StandardSpec<"stdc"> {
FunctionSpec<"frexpl", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntPtr>]>,
GuardedFunctionSpec<"frexpf128", RetValSpec<Float128Type>, [ArgSpec<Float128Type>, ArgSpec<IntPtr>], "LIBC_TYPES_HAS_FLOAT128">,
+ FunctionSpec<"fromfp", RetValSpec<DoubleType>, [ArgSpec<DoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpf", RetValSpec<FloatType>, [ArgSpec<FloatType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpl", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpf128", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+
+ FunctionSpec<"fromfpx", RetValSpec<DoubleType>, [ArgSpec<DoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpxf", RetValSpec<FloatType>, [ArgSpec<FloatType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpxl", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"fromfpxf128", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+
+ FunctionSpec<"ufromfp", RetValSpec<DoubleType>, [ArgSpec<DoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpf", RetValSpec<FloatType>, [ArgSpec<FloatType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpl", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpf128", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+
+ FunctionSpec<"ufromfpx", RetValSpec<DoubleType>, [ArgSpec<DoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpxf", RetValSpec<FloatType>, [ArgSpec<FloatType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpxl", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+ FunctionSpec<"ufromfpxf128", RetValSpec<LongDoubleType>, [ArgSpec<LongDoubleType>, ArgSpec<IntType>, ArgSpec<UnsignedIntType>]>,
+
FunctionSpec<"hypot", RetValSpec<DoubleType>, [ArgSpec<DoubleType>, ArgSpec<DoubleType>]>,
FunctionSpec<"hypotf", RetValSpec<FloatType>, [ArgSpec<FloatType>, ArgSpec<FloatType>]>,
diff --git a/libc/src/__support/FPUtil/NearestIntegerOperations.h b/libc/src/__support/FPUtil/NearestIntegerOperations.h
index e890e38ba4ae7d..cef612f9263529 100644
--- a/libc/src/__support/FPUtil/NearestIntegerOperations.h
+++ b/libc/src/__support/FPUtil/NearestIntegerOperations.h
@@ -141,7 +141,7 @@ LIBC_INLINE T round(T x) {
}
template <typename T, cpp::enable_if_t<cpp::is_floating_point_v<T>, int> = 0>
-LIBC_INLINE T round_using_current_rounding_mode(T x) {
+LIBC_INLINE constexpr T round_using_specific_rounding_mode(T x, int rnd) {
using StorageType = typename FPBits<T>::StorageType;
FPBits<T> bits(x);
@@ -151,7 +151,6 @@ LIBC_INLINE T round_using_current_rounding_mode(T x) {
bool is_neg = bits.is_neg();
int exponent = bits.get_exponent();
- int rounding_mode = quick_get_round();
// If the exponent is greater than the most negative mantissa
// exponent, then x is already an integer.
@@ -159,20 +158,23 @@ LIBC_INLINE T round_using_current_rounding_mode(T x) {
return x;
if (exponent <= -1) {
- switch (rounding_mode) {
- case FE_DOWNWARD:
+ switch (rnd) {
+ case FP_INT_DOWNWARD:
return is_neg ? T(-1.0) : T(0.0);
- case FE_UPWARD:
+ case FP_INT_UPWARD:
return is_neg ? T(-0.0) : T(1.0);
- case FE_TOWARDZERO:
+ case FP_INT_TOWARDZERO:
return is_neg ? T(-0.0) : T(0.0);
- case FE_TONEAREST:
+ case FP_INT_TONEARESTFROMZERO:
+ if (exponent < -1)
+ return is_neg ? T(-0.0) : T(0.0); // abs(x) < 0.5
+ return is_neg ? T(-1.0) : T(1.0); // abs(x) >= 0.5
+ case FP_INT_TONEAREST:
+ default:
if (exponent <= -2 || bits.get_mantissa() == 0)
return is_neg ? T(-0.0) : T(0.0); // abs(x) <= 0.5
else
return is_neg ? T(-1.0) : T(1.0); // abs(x) > 0.5
- default:
- __builtin_unreachable();
}
}
@@ -194,14 +196,19 @@ LIBC_INLINE T round_using_current_rounding_mode(T x) {
StorageType trunc_is_odd =
new_bits.get_mantissa() & (StorageType(1) << trim_size);
- switch (rounding_mode) {
- case FE_DOWNWARD:
+ switch (rnd) {
+ case FP_INT_DOWNWARD:
return is_neg ? trunc_value - T(1.0) : trunc_value;
- case FE_UPWARD:
+ case FP_INT_UPWARD:
return is_neg ? trunc_value : trunc_value + T(1.0);
- case FE_TOWARDZERO:
+ case FP_INT_TOWARDZERO:
return trunc_value;
- case FE_TONEAREST:
+ case FP_INT_TONEARESTFROMZERO:
+ if (trim_value >= half_value)
+ return is_neg ? trunc_value - T(1.0) : trunc_value + T(1.0);
+ return trunc_value;
+ case FP_INT_TONEAREST:
+ default:
if (trim_value > half_value) {
return is_neg ? trunc_value - T(1.0) : trunc_value + T(1.0);
} else if (trim_value == half_value) {
@@ -214,11 +221,66 @@ LIBC_INLINE T round_using_current_rounding_mode(T x) {
} else {
return trunc_value;
}
+ }
+}
+
+template <typename T, cpp::enable_if_t<cpp::is_floating_point_v<T>, int> = 0>
+LIBC_INLINE T round_using_current_rounding_mode(T x) {
+ int rounding_mode = quick_get_round();
+
+ switch (rounding_mode) {
+ case FE_DOWNWARD:
+ return round_using_specific_rounding_mode(x, FP_INT_DOWNWARD);
+ case FE_UPWARD:
+ return round_using_specific_rounding_mode(x, FP_INT_UPWARD);
+ case FE_TOWARDZERO:
+ return round_using_specific_rounding_mode(x, FP_INT_TOWARDZERO);
+ case FE_TONEAREST:
+ return round_using_specific_rounding_mode(x, FP_INT_TONEAREST);
default:
__builtin_unreachable();
}
}
+template <bool IsSigned, typename T,
+ cpp::enable_if_t<cpp::is_floating_point_v<T>, int> = 0>
+LIBC_INLINE constexpr T fromfp(T x, int rnd, unsigned int width) {
+ if (width == 0U)
+ return FPBits<T>::quiet_nan().get_val();
+
+ T rounded_value = round_using_specific_rounding_mode(x, rnd);
+
+ if constexpr (IsSigned) {
+ // T can't hold a finite number >= 2.0 * 2^EXP_BIAS.
+ if (width - 1 > FPBits<T>::EXP_BIAS)
+ return rounded_value;
+ if (rounded_value < -T(1U << (width - 1U)))
+ return FPBits<T>::quiet_nan().get_val();
+ if (rounded_value > T((1U << (width - 1U)) - 1U))
+ return FPBits<T>::quiet_nan().get_val();
+ return rounded_value;
+ }
+
+ if (rounded_value < T(0.0))
+ return FPBits<T>::quiet_nan().get_val();
+ // T can't hold a finite number >= 2.0 * 2^EXP_BIAS.
+ if (width <= FPBits<T>::EXP_BIAS && rounded_value > T(1U << width) - 1U)
+ return FPBits<T>::quiet_nan().get_val();
+ return rounded_value;
+}
+
+template <bool IsSigned, typename T,
+ cpp::enable_if_t<cpp::is_floating_point_v<T>, int> = 0>
+LIBC_INLINE constexpr T fromfpx(T x, int rnd, unsigned int width) {
+ T rounded_value = fromfp<IsSigned>(x, rnd, width);
+ FPBits<T> bits(rounded_value);
+
+ if (!bits.is_nan() && rounded_value != x)
+ raise_except_if_required(FE_INEXACT);
+
+ return rounded_value;
+}
+
namespace internal {
template <typename F, typename I,
diff --git a/libc/src/math/CMakeLists.txt b/libc/src/math/CMakeLists.txt
index 5e2e6e699d0e0c..2f1d65780013f8 100644
--- a/libc/src/math/CMakeLists.txt
+++ b/libc/src/math/CMakeLists.txt
@@ -127,6 +127,16 @@ add_math_entrypoint_object(frexpf)
add_math_entrypoint_object(frexpl)
add_math_entrypoint_object(frexpf128)
+add_math_entrypoint_object(fromfp)
+add_math_entrypoint_object(fromfpf)
+add_math_entrypoint_object(fromfpl)
+add_math_entrypoint_object(fromfpf128)
+
+add_math_entrypoint_object(fromfpx)
+add_math_entrypoint_object(fromfpxf)
+add_math_entrypoint_object(fromfpxl)
+add_math_entrypoint_object(fromfpxf128)
+
add_math_entrypoint_object(hypot)
add_math_entrypoint_object(hypotf)
@@ -267,3 +277,13 @@ add_math_entrypoint_object(trunc)
add_math_entrypoint_object(truncf)
add_math_entrypoint_object(truncl)
add_math_entrypoint_object(truncf128)
+
+add_math_entrypoint_object(ufromfp)
+add_math_entrypoint_object(ufromfpf)
+add_math_entrypoint_object(ufromfpl)
+add_math_entrypoint_object(ufromfpf128)
+
+add_math_entrypoint_object(ufromfpx)
+add_math_entrypoint_object(ufromfpxf)
+add_math_entrypoint_object(ufromfpxl)
+add_math_entrypoint_object(ufromfpxf128)
diff --git a/libc/src/math/fromfp.h b/libc/src/math/fromfp.h
new file mode 100644
index 00000000000000..d3de2dd3460832
--- /dev/null
+++ b/libc/src/math/fromfp.h
@@ -0,0 +1,18 @@
+//===-- Implementation header for fromfp ------------------------*- 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_FROMFP_H
+#define LLVM_LIBC_SRC_MATH_FROMFP_H
+
+namespace LIBC_NAMESPACE {
+
+double fromfp(double x, int rnd, unsigned int width);
+
+} // namespace LIBC_NAMESPACE
+
+#endif // LLVM_LIBC_SRC_MATH_FROMFP_H
diff --git a/libc/src/math/fromfpf.h b/libc/src/math/fromfpf.h
new file mode 100644
index 00000000000000..11d432148d0189
--- /dev/null
+++ b/libc/src/math/fromfpf.h
@@ -0,0 +1,18 @@
+//===-- Implementation header for fromfpf -----------------------*- 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_FROMFPF_H
+#define LLVM_LIBC_SRC_MATH_FROMFPF_H
+
+namespace LIBC_NAMESPACE {
+
+float fromfpf(float x, int rnd, unsigned int width);
+
+} // namespace LIBC_NAMESPACE
+
+#endif // LLVM_LIBC_SRC_MATH_FROMFPF_H
diff --git a/libc/src/math/fromfpf128.h b/libc/src/math/fromfpf128.h
new file mode 100644
index 00000000000000..5f85fde570a0fb
--- /dev/null
+++ b/libc/src/math/fromfpf128.h
@@ -0,0 +1,20 @@
+//===-- Implementation header for fromfpf128 --------------------*- 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_FROMFPF128_H
+#define LLVM_LIBC_SRC_MATH_FROMFPF128_H
+
+#include "src/__support/macros/properties/types.h"
+
+namespace LIBC_NAMESPACE {
+
+float128 fromfpf128(float128 x, int rnd, unsigned int width);
+
+} // namespace LIBC_NAMESPACE
+
+#endif // LLVM_LIBC_SRC_MATH_FROMFPF128_H
diff --git a/libc/src/math/fromfpl.h b/libc/src/math/fromfpl.h
new file mode 100644
index 00000000000000..dd8e1eebdea994
--- /dev/null
+++ b/libc/src/math/fromfpl.h
@@ -0,0 +1,18 @@
+//===-- Implementation header for fromfpl -----------------------*- 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_FROMFPL_H
+#define LLVM_LIBC_SRC_MATH_FROMFPL_H
+
+namespace LIBC_NAMESPACE {
+
+long double fromfpl(long double x, int rnd, unsigned int width);
+
+} // namespace LIBC_NAMESPACE
+
+#endif // LLVM_LIBC_SRC_MATH_FROMFPL_H
diff --git a/libc/src/math/fromfpx.h b/libc/src/math/fromfpx.h
new file mode 100644
index 0000000000...
[truncated]
|
overmighty
commented
Mar 20, 2024
overmighty
commented
Mar 20, 2024
| HDRS | ||
| FromfpTest.h | ||
| DEPENDS | ||
| libc.src.math.fromfp |
There was a problem hiding this comment.
In my previous PR for nextup and nextdown I had added more dependencies: https://github.com/llvm/llvm-project/pull/85431/files#diff-2269fc16d8717f30b774875846f93f54879dae7515bdfd3dd5d5645fd7413aa7R1626-R1628.
Now that I think about it again, I assume that depending on libc.src.math.<func> is enough to also depend on the header and dependencies declared in the add_entrypoint_object call for the function in libc/src/math/generic/CMakeLists.txt.
There was a problem hiding this comment.
Yes, but include-what-you-use is always better, instead of completely relying on transitive dependence. So if your test directly calls something in those targets, it's always better to directly include them in our DEPENDS list.
There was a problem hiding this comment.
Right, but the nextup and nextdown tests only directly use libc.src.math.{nextup,nextdown} and test/UnitTest/{FPMatcher.h,Test.h}, and add_fp_unittest already takes care of the latter.
overmighty
commented
Mar 20, 2024
overmighty
commented
Mar 20, 2024
| EXPECT_FP_EQ(aNaN, func(T(-1234.96), FP_INT_DOWNWARD, 11U)); | ||
| } | ||
|
|
||
| void testFractionsTowardZeroWithinRange(FromfpxFunc func) { |
There was a problem hiding this comment.
The formatting for the tests that use EXPECT_FP_EQ_WITH_EXCEPTION isn't great with lines of different lengths triggering different wrap-arounds, but I'm not sure if it's a problem.
Note that FromfpxTest.h is pretty much a copy-paste of FromfpTest.h with testFractions*WithinRange tests changed to use EXPECT_FP_EQ_WITH_EXCEPTION. Same for UfromfpxTest.h vs UfromfpTest.h. I'm not sure if something should be done about this either.
lntue
reviewed
Mar 20, 2024
lntue
reviewed
Mar 20, 2024
lntue
reviewed
Mar 20, 2024
lntue
reviewed
Mar 20, 2024
|
I added tests for the fallback to |
lntue
reviewed
Mar 22, 2024
libc/docs/dev/undefined_behavior.rst
Outdated
|
|
||
| Unknown Math Rounding Direction | ||
| ------------------------------- | ||
| The C standard states that if the value of the ``rnd`` argument of the |
lntue
reviewed
Mar 22, 2024
libc/src/math/generic/CMakeLists.txt
Outdated
| libc.src.__support.macros.properties.types | ||
| libc.src.__support.FPUtil.nearest_integer | ||
| COMPILE_OPTIONS | ||
| -O3 |
overmighty
force-pushed
the
libc-math-fromfp-fromfpx-ufromfp-ufromfpx
branch
from
9616f21 to
ac71b02
Compare
|
Had to rebase as Bazel builds were broken before #86298. |
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
|
✅ With the latest revision this PR passed the Python code formatter. |
lntue
reviewed
Mar 24, 2024
| @@ -141,7 +141,7 @@ LIBC_INLINE T round(T x) { | |||
| } | |||
|
|
|||
| template <typename T, cpp::enable_if_t<cpp::is_floating_point_v<T>, int> = 0> | |||
| LIBC_INLINE T round_using_current_rounding_mode(T x) { | |||
| LIBC_INLINE constexpr T round_using_specific_rounding_mode(T x, int rnd) { | |||
There was a problem hiding this comment.
Since you are here, would you mind updating the enable_if to be on the returning type instead:
template <typename T>
LIBC_INLINE constexpr typename cpp::enable_if_t<..., T> rounding_using_specific_rounding_mode() { ... }
Same for the other 2 functions below.
There was a problem hiding this comment.
CLion Nova warns about the typename keyword before cpp::enable_if_t being redundant.
There was a problem hiding this comment.
You can remove typename before cpp::enable_if_t if it's not needed.
lntue
approved these changes
Mar 25, 2024
|
Thanks for adding these functions! I'll merge the patch in the morning when I can watch the bots. |
overmighty
commented
Mar 25, 2024
| LIBC_INLINE T round_using_current_rounding_mode(T x) { | ||
| template <typename T> | ||
| LIBC_INLINE constexpr cpp::enable_if_t<cpp::is_floating_point_v<T>, T> | ||
| round_using_specific_rounding_mode(T x, int rnd) { |
There was a problem hiding this comment.
Maybe this function should have been made static or put in a namespace.
|
You're welcome. By the way, |
Addressed by #87678. |
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.
Fixes #85279.
cc @lntue