[libc-commits] [libc] [llvm] [libc][math] Refactor logf16 to header-only shared math (PR #175408)

[cpist via libc-commits]

================
@@ -611,6 +611,33 @@ add_header_library(
     libc.src.__support.FPUtil.manipulation_functions
 )
 
+add_header_library(
+  f16sqrtl
+  HDRS
+    f16sqrtl.h
+  DEPENDS
+    libc.src.__support.FPUtil.generic.sqrt
+    libc.src.__support.macros.properties.types
+)
+
+add_header_library(
+  frexpf128
+  HDRS
+    frexpf128.h
+  DEPENDS
+    libc.src.__support.macros.properties.types
+    libc.src.__support.FPUtil.manipulation_functions
+)
+
+add_header_library(
+  frexpf128
+  HDRS
+    frexpf128.h
+  DEPENDS
+    libc.src.__support.macros.properties.types
+    libc.src.__support.FPUtil.manipulation_functions
+)
+
----------------
cpist wrote:

I think I made a mistake while rushing through the rebase this morning.
I've checked CMakeLists.txt and re-uploaded the changes.

https://github.com/llvm/llvm-project/pull/175408