From 3f76cdfe25a02df1a4371edae1eb16555be5260a Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Sat, 5 Apr 2025 12:20:55 -0400
Subject: [PATCH 1/6] [libc] Extend fputil::sqrt to use floating point
 instruction for arm32 CPUs if FPUs are available.

---
 libc/src/__support/FPUtil/aarch64/sqrt.h       | 11 ++++++++---
 libc/src/__support/FPUtil/sqrt.h               |  2 +-
 .../__support/macros/properties/cpu_features.h | 18 +++++++++++++++---
 3 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/libc/src/__support/FPUtil/aarch64/sqrt.h b/libc/src/__support/FPUtil/aarch64/sqrt.h
index b69267ff91f5c..cfd3579f621d0 100644
--- a/libc/src/__support/FPUtil/aarch64/sqrt.h
+++ b/libc/src/__support/FPUtil/aarch64/sqrt.h
@@ -12,8 +12,9 @@
 #include "src/__support/common.h"
 #include "src/__support/macros/config.h"
 #include "src/__support/macros/properties/architectures.h"
+#include "src/__support/macros/properties/cpu_features.h"
 
-#if !defined(LIBC_TARGET_ARCH_IS_AARCH64)
+#if !defined(LIBC_TARGET_ARCH_IS_ANY_ARM)
 #error "Invalid include"
 #endif
 
@@ -22,17 +23,21 @@
 namespace LIBC_NAMESPACE_DECL {
 namespace fputil {
 
+#ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
 template <> LIBC_INLINE float sqrt<float>(float x) {
   float y;
-  __asm__ __volatile__("fsqrt %s0, %s1\n\t" : "=w"(y) : "w"(x));
+  asm("fsqrt %s0, %s1\n\t" : "=w"(y) : "w"(x));
   return y;
 }
+#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
 
+#ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
 template <> LIBC_INLINE double sqrt<double>(double x) {
   double y;
-  __asm__ __volatile__("fsqrt %d0, %d1\n\t" : "=w"(y) : "w"(x));
+  asm("fsqrt %d0, %d1\n\t" : "=w"(y) : "w"(x));
   return y;
 }
+#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
 
 } // namespace fputil
 } // namespace LIBC_NAMESPACE_DECL
diff --git a/libc/src/__support/FPUtil/sqrt.h b/libc/src/__support/FPUtil/sqrt.h
index eb86ddfa89d8e..ca9890600168f 100644
--- a/libc/src/__support/FPUtil/sqrt.h
+++ b/libc/src/__support/FPUtil/sqrt.h
@@ -14,7 +14,7 @@
 
 #if defined(LIBC_TARGET_ARCH_IS_X86_64) && defined(LIBC_TARGET_CPU_HAS_SSE2)
 #include "x86_64/sqrt.h"
-#elif defined(LIBC_TARGET_ARCH_IS_AARCH64)
+#elif defined(LIBC_TARGET_ARCH_IS_ANY_ARM)
 #include "aarch64/sqrt.h"
 #elif defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
 #include "riscv/sqrt.h"
diff --git a/libc/src/__support/macros/properties/cpu_features.h b/libc/src/__support/macros/properties/cpu_features.h
index 1714775ca334d..47dd8a03613e7 100644
--- a/libc/src/__support/macros/properties/cpu_features.h
+++ b/libc/src/__support/macros/properties/cpu_features.h
@@ -42,18 +42,30 @@
 #define LIBC_TARGET_CPU_HAS_AVX512BW
 #endif
 
+#if defined(__ARM_FP)
+#if (__ARM_FP & 0x2)
+#define LIBC_TARGET_CPU_HAS_ARM_FPU_HALF
+#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_HALF
+#if (__ARM_FP & 0x4)
+#define LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
+#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
+#if (__ARM_FP & 0x8)
+#define LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
+#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
+#endif // __ARM_FP
+
 #if defined(__ARM_FEATURE_FMA) || (defined(__AVX2__) && defined(__FMA__)) ||   \
     defined(__NVPTX__) || defined(__AMDGPU__) || defined(__LIBC_RISCV_USE_FMA)
 #define LIBC_TARGET_CPU_HAS_FMA
 // Provide a more fine-grained control of FMA instruction for ARM targets.
 #if defined(__ARM_FP)
-#if (__ARM_FP & 0x2)
+#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_HALF)
 #define LIBC_TARGET_CPU_HAS_FMA_HALF
 #endif // LIBC_TARGET_CPU_HAS_FMA_HALF
-#if (__ARM_FP & 0x4)
+#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT)
 #define LIBC_TARGET_CPU_HAS_FMA_FLOAT
 #endif // LIBC_TARGET_CPU_HAS_FMA_FLOAT
-#if (__ARM_FP & 0x8)
+#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE)
 #define LIBC_TARGET_CPU_HAS_FMA_DOUBLE
 #endif // LIBC_TARGET_CPU_HAS_FMA_DOUBLE
 #else

From 3be70f13494be4eb5c8a6899ae2094a315c259d4 Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Sat, 5 Apr 2025 14:11:50 -0400
Subject: [PATCH 2/6] Use __builtin_elementwise_sqrt if available.

---
 libc/src/__support/FPUtil/aarch64/sqrt.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/libc/src/__support/FPUtil/aarch64/sqrt.h b/libc/src/__support/FPUtil/aarch64/sqrt.h
index cfd3579f621d0..9462fc16f3f4c 100644
--- a/libc/src/__support/FPUtil/aarch64/sqrt.h
+++ b/libc/src/__support/FPUtil/aarch64/sqrt.h
@@ -25,17 +25,25 @@ namespace fputil {
 
 #ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
 template <> LIBC_INLINE float sqrt<float>(float x) {
+#if __has_builtin(__builtin_elementwise_sqrt)
+  return __builtin_elementwise_sqrt(x);
+#else
   float y;
   asm("fsqrt %s0, %s1\n\t" : "=w"(y) : "w"(x));
   return y;
+#endif // __builtin_elementwise_sqrt
 }
 #endif // LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
 
 #ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
 template <> LIBC_INLINE double sqrt<double>(double x) {
+#if __has_builtin(__builtin_elementwise_sqrt)
+  return __builtin_elementwise_sqrt(x);
+#else
   double y;
   asm("fsqrt %d0, %d1\n\t" : "=w"(y) : "w"(x));
   return y;
+#endif // __builtin_elementwise_sqrt
 }
 #endif // LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
 

From b97174a320c43d131faa341871106354daad9ccd Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Sun, 6 Apr 2025 12:10:11 -0400
Subject: [PATCH 3/6] Move __builtin_elementwise_sqrt to generic, and add GPUs,
 RISCV support.

---
 .../__support/FPUtil/{aarch64 => arm}/sqrt.h  | 24 ++++--------
 libc/src/__support/FPUtil/riscv/sqrt.h        | 11 +++---
 libc/src/__support/FPUtil/sqrt.h              | 37 +++++++++++++++++--
 libc/src/__support/FPUtil/x86_64/sqrt.h       |  2 -
 .../macros/properties/cpu_features.h          | 37 +++++++++++++++----
 .../llvm-project-overlay/libc/BUILD.bazel     |  3 +-
 6 files changed, 76 insertions(+), 38 deletions(-)
 rename libc/src/__support/FPUtil/{aarch64 => arm}/sqrt.h (61%)

diff --git a/libc/src/__support/FPUtil/aarch64/sqrt.h b/libc/src/__support/FPUtil/arm/sqrt.h
similarity index 61%
rename from libc/src/__support/FPUtil/aarch64/sqrt.h
rename to libc/src/__support/FPUtil/arm/sqrt.h
index 9462fc16f3f4c..39ac5395f869e 100644
--- a/libc/src/__support/FPUtil/aarch64/sqrt.h
+++ b/libc/src/__support/FPUtil/arm/sqrt.h
@@ -6,8 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
-#ifndef LLVM_LIBC_SRC___SUPPORT_FPUTIL_AARCH64_SQRT_H
-#define LLVM_LIBC_SRC___SUPPORT_FPUTIL_AARCH64_SQRT_H
+#ifndef LLVM_LIBC_SRC___SUPPORT_FPUTIL_ARM_SQRT_H
+#define LLVM_LIBC_SRC___SUPPORT_FPUTIL_ARM_SQRT_H
 
 #include "src/__support/common.h"
 #include "src/__support/macros/config.h"
@@ -18,36 +18,26 @@
 #error "Invalid include"
 #endif
 
-#include "src/__support/FPUtil/generic/sqrt.h"
-
 namespace LIBC_NAMESPACE_DECL {
 namespace fputil {
 
-#ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
+#ifdef LIBC_TARGET_CPU_HAS_FPU_FLOAT
 template <> LIBC_INLINE float sqrt<float>(float x) {
-#if __has_builtin(__builtin_elementwise_sqrt)
-  return __builtin_elementwise_sqrt(x);
-#else
   float y;
   asm("fsqrt %s0, %s1\n\t" : "=w"(y) : "w"(x));
   return y;
-#endif // __builtin_elementwise_sqrt
 }
-#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
+#endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
 
-#ifdef LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
+#ifdef LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 template <> LIBC_INLINE double sqrt<double>(double x) {
-#if __has_builtin(__builtin_elementwise_sqrt)
-  return __builtin_elementwise_sqrt(x);
-#else
   double y;
   asm("fsqrt %d0, %d1\n\t" : "=w"(y) : "w"(x));
   return y;
-#endif // __builtin_elementwise_sqrt
 }
-#endif // LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
+#endif // LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 
 } // namespace fputil
 } // namespace LIBC_NAMESPACE_DECL
 
-#endif // LLVM_LIBC_SRC___SUPPORT_FPUTIL_AARCH64_SQRT_H
+#endif // LLVM_LIBC_SRC___SUPPORT_FPUTIL_ARM_SQRT_H
diff --git a/libc/src/__support/FPUtil/riscv/sqrt.h b/libc/src/__support/FPUtil/riscv/sqrt.h
index 0363822a4e8af..d5d0d9bb5111e 100644
--- a/libc/src/__support/FPUtil/riscv/sqrt.h
+++ b/libc/src/__support/FPUtil/riscv/sqrt.h
@@ -12,31 +12,30 @@
 #include "src/__support/common.h"
 #include "src/__support/macros/config.h"
 #include "src/__support/macros/properties/architectures.h"
+#include "src/__support/macros/properties/cpu_features.h"
 
 #if !defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
 #error "Invalid include"
 #endif
 
-#include "src/__support/FPUtil/generic/sqrt.h"
-
 namespace LIBC_NAMESPACE_DECL {
 namespace fputil {
 
-#ifdef __riscv_flen
+#ifdef LIBC_TARGET_CPU_HAS_FPU_FLOAT
 template <> LIBC_INLINE float sqrt<float>(float x) {
   float result;
   __asm__ __volatile__("fsqrt.s %0, %1\n\t" : "=f"(result) : "f"(x));
   return result;
 }
+#endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
 
-#if __riscv_flen >= 64
+#if LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 template <> LIBC_INLINE double sqrt<double>(double x) {
   double result;
   __asm__ __volatile__("fsqrt.d %0, %1\n\t" : "=f"(result) : "f"(x));
   return result;
 }
-#endif // __riscv_flen >= 64
-#endif // __riscv_flen
+#endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
 
 } // namespace fputil
 } // namespace LIBC_NAMESPACE_DECL
diff --git a/libc/src/__support/FPUtil/sqrt.h b/libc/src/__support/FPUtil/sqrt.h
index ca9890600168f..4819935b8b667 100644
--- a/libc/src/__support/FPUtil/sqrt.h
+++ b/libc/src/__support/FPUtil/sqrt.h
@@ -12,14 +12,43 @@
 #include "src/__support/macros/properties/architectures.h"
 #include "src/__support/macros/properties/cpu_features.h"
 
-#if defined(LIBC_TARGET_ARCH_IS_X86_64) && defined(LIBC_TARGET_CPU_HAS_SSE2)
+#include "src/__support/FPUtil/generic/sqrt.h"
+
+// Generic instruction specializations with __builtin_elementwise_sqrt.
+#if defined(LIBC_TARGET_CPU_HAS_FPU_FLOAT) ||                                  \
+    defined(LIBC_TARGET_CPU_HAS_FPU_DOUBLE)
+
+#if __has_builtin(__builtin_elementwise_sqrt)
+
+namespace LIBC_NAMESPACE_DECL {
+namespace fputil {
+
+#ifdef LIBC_TARGET_CPU_HAS_FPU_FLOAT
+template <> LIBC_INLINE float sqrt<float>(float x) {
+  return __builtin_elementwise_sqrt(x);
+}
+#endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
+
+#ifdef LIBC_TARGET_CPU_HAS_FPU_DOUBLE
+template <> LIBC_INLINE double sqrt<double>(double x) {
+  return __builtin_elementwise_sqrt(x);
+}
+#endif // LIBC_TARGET_CPU_HAS_FPU_DOUBLE
+
+} // namespace fputil
+} // namespace LIBC_NAMESPACE_DECL
+
+#else
+// Use inline assembly when __builtin_elementwise_sqrt is not available.
+#if defined(LIBC_TARGET_CPU_HAS_SSE2)
 #include "x86_64/sqrt.h"
 #elif defined(LIBC_TARGET_ARCH_IS_ANY_ARM)
-#include "aarch64/sqrt.h"
+#include "arm/sqrt.h"
 #elif defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
 #include "riscv/sqrt.h"
-#else
-#include "generic/sqrt.h"
+
+#endif // __builtin_elementwise_sqrt
 
 #endif
+
 #endif // LLVM_LIBC_SRC___SUPPORT_FPUTIL_SQRT_H
diff --git a/libc/src/__support/FPUtil/x86_64/sqrt.h b/libc/src/__support/FPUtil/x86_64/sqrt.h
index e10447751d216..eae40cc8829f1 100644
--- a/libc/src/__support/FPUtil/x86_64/sqrt.h
+++ b/libc/src/__support/FPUtil/x86_64/sqrt.h
@@ -18,8 +18,6 @@
 #error "sqrtss / sqrtsd need SSE2"
 #endif
 
-#include "src/__support/FPUtil/generic/sqrt.h"
-
 namespace LIBC_NAMESPACE_DECL {
 namespace fputil {
 
diff --git a/libc/src/__support/macros/properties/cpu_features.h b/libc/src/__support/macros/properties/cpu_features.h
index 47dd8a03613e7..bdc08e79f173e 100644
--- a/libc/src/__support/macros/properties/cpu_features.h
+++ b/libc/src/__support/macros/properties/cpu_features.h
@@ -20,6 +20,8 @@
 
 #if defined(__SSE2__)
 #define LIBC_TARGET_CPU_HAS_SSE2
+#define LIBC_TARGET_CPU_HAS_FPU_FLOAT
+#define LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 #endif
 
 #if defined(__SSE4_2__)
@@ -45,33 +47,52 @@
 #if defined(__ARM_FP)
 #if (__ARM_FP & 0x2)
 #define LIBC_TARGET_CPU_HAS_ARM_FPU_HALF
+#define LIBC_TARGET_CPU_HAS_FPU_HALF
 #endif // LIBC_TARGET_CPU_HAS_ARM_FPU_HALF
 #if (__ARM_FP & 0x4)
 #define LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
+#define LIBC_TARGET_CPU_HAS_FPU_FLOAT
 #endif // LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT
 #if (__ARM_FP & 0x8)
 #define LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
+#define LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 #endif // LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE
 #endif // __ARM_FP
 
+#if defined(__riscv_flen)
+// https://github.com/riscv-non-isa/riscv-c-api-doc/blob/main/src/c-api.adoc
+#if (__riscv_flen & 0x20)
+#define LIBC_TARGET_CPU_HAS_RISCV_FPU_HALF
+#define LIBC_TARGET_CPU_HAS_FPU_HALF
+#endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_HALF
+#if (__riscv_flen & 0x40)
+#define LIBC_TARGET_CPU_HAS_RISCV_FPU_FLOAT
+#define LIBC_TARGET_CPU_HAS_FPU_FLOAT
+#endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_FLOAT
+#if (__riscv_flen & 0x80)
+#define LIBC_TARGET_CPU_HAS_RISCV_FPU_DOUBLE
+#define LIBC_TARGET_CPU_HAS_FPU_DOUBLE
+#endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_DOUBLE
+#endif // __riscv_flen
+
+#if defined(__NVPTX__) || defined(__AMDGPU__)
+#define LIBC_TARGET_CPU_HAS_FPU_FLOAT
+#define LIBC_TARGET_CPU_HAS_FPU_DOUBLE
+#endif
+
 #if defined(__ARM_FEATURE_FMA) || (defined(__AVX2__) && defined(__FMA__)) ||   \
     defined(__NVPTX__) || defined(__AMDGPU__) || defined(__LIBC_RISCV_USE_FMA)
 #define LIBC_TARGET_CPU_HAS_FMA
 // Provide a more fine-grained control of FMA instruction for ARM targets.
-#if defined(__ARM_FP)
-#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_HALF)
+#if defined(LIBC_TARGET_CPU_HAS_FPU_HALF)
 #define LIBC_TARGET_CPU_HAS_FMA_HALF
 #endif // LIBC_TARGET_CPU_HAS_FMA_HALF
-#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_FLOAT)
+#if defined(LIBC_TARGET_CPU_HAS_FPU_FLOAT)
 #define LIBC_TARGET_CPU_HAS_FMA_FLOAT
 #endif // LIBC_TARGET_CPU_HAS_FMA_FLOAT
-#if defined(LIBC_TARGET_CPU_HAS_ARM_FPU_DOUBLE)
+#if defined(LIBC_TARGET_CPU_HAS_FPU_DOUBLE)
 #define LIBC_TARGET_CPU_HAS_FMA_DOUBLE
 #endif // LIBC_TARGET_CPU_HAS_FMA_DOUBLE
-#else
-#define LIBC_TARGET_CPU_HAS_FMA_FLOAT
-#define LIBC_TARGET_CPU_HAS_FMA_DOUBLE
-#endif
 #endif
 
 #if defined(LIBC_TARGET_ARCH_IS_AARCH64) ||                                    \
diff --git a/utils/bazel/llvm-project-overlay/libc/BUILD.bazel b/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
index a9357485d3a10..dcb4d53f9dad0 100644
--- a/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
@@ -1179,7 +1179,7 @@ sqrt_hdrs = selects.with_or({
         "src/__support/FPUtil/x86_64/sqrt.h",
     ],
     PLATFORM_CPU_ARM64: sqrt_common_hdrs + [
-        "src/__support/FPUtil/aarch64/sqrt.h",
+        "src/__support/FPUtil/arm/sqrt.h",
     ],
 })
 
@@ -1195,6 +1195,7 @@ libc_support_library(
         ":__support_fputil_fenv_impl",
         ":__support_fputil_fp_bits",
         ":__support_fputil_rounding_mode",
+        ":__support_macros_properties_cpu_features",
         ":__support_uint128",
     ],
 )

From 0590dbad8579ebf12f41c11637e2a4d2cbfd1d2b Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Sun, 6 Apr 2025 12:15:29 -0400
Subject: [PATCH 4/6] Add missing #endif.

---
 libc/src/__support/FPUtil/sqrt.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/libc/src/__support/FPUtil/sqrt.h b/libc/src/__support/FPUtil/sqrt.h
index 4819935b8b667..bf0a99ff94225 100644
--- a/libc/src/__support/FPUtil/sqrt.h
+++ b/libc/src/__support/FPUtil/sqrt.h
@@ -46,9 +46,10 @@ template <> LIBC_INLINE double sqrt<double>(double x) {
 #include "arm/sqrt.h"
 #elif defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
 #include "riscv/sqrt.h"
+#endif // Target specific header of inline asm.
 
 #endif // __builtin_elementwise_sqrt
 
-#endif
+#endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT or DOUBLE
 
 #endif // LLVM_LIBC_SRC___SUPPORT_FPUTIL_SQRT_H

From 70927af8c721097881a11e722f69be51dfad124c Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Sun, 6 Apr 2025 12:24:32 -0400
Subject: [PATCH 5/6] Fix __riscv_flen mask.

---
 libc/src/__support/macros/properties/cpu_features.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/libc/src/__support/macros/properties/cpu_features.h b/libc/src/__support/macros/properties/cpu_features.h
index bdc08e79f173e..3677e1fc3275c 100644
--- a/libc/src/__support/macros/properties/cpu_features.h
+++ b/libc/src/__support/macros/properties/cpu_features.h
@@ -61,15 +61,15 @@
 
 #if defined(__riscv_flen)
 // https://github.com/riscv-non-isa/riscv-c-api-doc/blob/main/src/c-api.adoc
-#if (__riscv_flen & 0x20)
+#if (__riscv_flen & 0x10)
 #define LIBC_TARGET_CPU_HAS_RISCV_FPU_HALF
 #define LIBC_TARGET_CPU_HAS_FPU_HALF
 #endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_HALF
-#if (__riscv_flen & 0x40)
+#if (__riscv_flen & 0x20)
 #define LIBC_TARGET_CPU_HAS_RISCV_FPU_FLOAT
 #define LIBC_TARGET_CPU_HAS_FPU_FLOAT
 #endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_FLOAT
-#if (__riscv_flen & 0x80)
+#if (__riscv_flen & 0x40)
 #define LIBC_TARGET_CPU_HAS_RISCV_FPU_DOUBLE
 #define LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 #endif // LIBC_TARGET_CPU_HAS_RISCV_FPU_DOUBLE

From 51014dd71fd2bee31422560a648bc6338b9656b0 Mon Sep 17 00:00:00 2001
From: Tue Ly <lntue@google.com>
Date: Tue, 8 Apr 2025 19:17:43 -0400
Subject: [PATCH 6/6] Address comments.

---
 libc/src/__support/FPUtil/riscv/sqrt.h | 4 ++--
 libc/src/__support/FPUtil/sqrt.h       | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/libc/src/__support/FPUtil/riscv/sqrt.h b/libc/src/__support/FPUtil/riscv/sqrt.h
index d5d0d9bb5111e..694451d02e2b1 100644
--- a/libc/src/__support/FPUtil/riscv/sqrt.h
+++ b/libc/src/__support/FPUtil/riscv/sqrt.h
@@ -24,7 +24,7 @@ namespace fputil {
 #ifdef LIBC_TARGET_CPU_HAS_FPU_FLOAT
 template <> LIBC_INLINE float sqrt<float>(float x) {
   float result;
-  __asm__ __volatile__("fsqrt.s %0, %1\n\t" : "=f"(result) : "f"(x));
+  asm("fsqrt.s %0, %1\n\t" : "=f"(result) : "f"(x));
   return result;
 }
 #endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
@@ -32,7 +32,7 @@ template <> LIBC_INLINE float sqrt<float>(float x) {
 #if LIBC_TARGET_CPU_HAS_FPU_DOUBLE
 template <> LIBC_INLINE double sqrt<double>(double x) {
   double result;
-  __asm__ __volatile__("fsqrt.d %0, %1\n\t" : "=f"(result) : "f"(x));
+  asm("fsqrt.d %0, %1\n\t" : "=f"(result) : "f"(x));
   return result;
 }
 #endif // LIBC_TARGET_CPU_HAS_FPU_FLOAT
diff --git a/libc/src/__support/FPUtil/sqrt.h b/libc/src/__support/FPUtil/sqrt.h
index bf0a99ff94225..9b151c4c5e1b3 100644
--- a/libc/src/__support/FPUtil/sqrt.h
+++ b/libc/src/__support/FPUtil/sqrt.h
@@ -38,7 +38,7 @@ template <> LIBC_INLINE double sqrt<double>(double x) {
 } // namespace fputil
 } // namespace LIBC_NAMESPACE_DECL
 
-#else
+#else // __builtin_elementwise_sqrt
 // Use inline assembly when __builtin_elementwise_sqrt is not available.
 #if defined(LIBC_TARGET_CPU_HAS_SSE2)
 #include "x86_64/sqrt.h"