[libc-commits] [PATCH] D150399: [libc] add options to printf decimal floats

[Tue Ly via Phabricator via libc-commits]

lntue added inline comments.


================
Comment at: libc/src/__support/FPUtil/dyadic_float.h:128
+    if (exponent > 0) {
+      new_mant = new_mant << exponent;
+    } else {
----------------
Use in-place `<<=` or add it to `UInt` if missing.


================
Comment at: libc/src/__support/FPUtil/dyadic_float.h:130
+    } else {
+      new_mant = new_mant >> (-exponent);
+    }
----------------
Use in-place `>>=` or add it to `UInt` if missing.


================
Comment at: libc/src/__support/float_to_string.h:144
+  constexpr cpp::UInt<INT_SIZE> MOD_SIZE =
+      (cpp::UInt<INT_SIZE>(1000000000)
+       << (CALC_SHIFT_CONST + (IDX_SIZE > 1 ? IDX_SIZE : 0)));
----------------
I'm just curious, will changing from `1000000000 * 2^(...)` to `1953125 * 2^(... + 9)` has any affect, like performance or accuracy?  Same with other instances below.


================
Comment at: libc/src/__support/float_to_string.h:347
+
+  return (val % 1000000000)[0];
+  // return fast_uint_mod_1e9(val);
----------------
Use `val.div_uint32_times_pow_2(1000000000, 0)` to speed up?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D150399/new/

https://reviews.llvm.org/D150399