From d11849bc0bf45452e6891c87c7c4ccdd7e541b9f Mon Sep 17 00:00:00 2001
From: Khalil Estell <kammcecorp@gmail.com>
Date: Sun, 19 Jul 2020 13:09:10 -0700
Subject: [PATCH] Add FMT_REDUCE_INT_INSTANTIATIONS flag (#1781)

* Remove <typename UInt> from int_writer

Reduce code bloat by removing multiple instantiation of int_writer based
on the <typename UInt> parameter.

Rationale:
- The only functions that gains a speedup by int size would be
  int_writer::on_dec()'s call to count_digits which uses CLZ. Thus to
  still take advantage of this speedup, we store the size of the int
  so we can use a switch statement to call the correct count_digits.
- All other implementations of count_digits require some sort of looping
  that terminates when the value hits zero regardless of what sized int
  it is.

Caveats:
- There is a performance hit when dealing with and passing around
  64-bit/128-bit values compared to 32-bit values on 32-bit platforms,
  and with 64-bit values on 64-bit systems. But this should not reduce the
  performance that dramatically.
- There is also a performance hit for on_dec() due to the addition of a
  switch case. But, due to it size, this should reduce to a jump table.

Resolves #1778

* Add FMT_USE_SMALLEST_INT flag

When defined and set to zero, will use the largest available integer
container for writing ints. The has the benefit of reducing instances
the of int_writer class which will reduce the binary cost.

* Rename flag to FMT_REDUCE_INT_INSTANTIATIONS

Add comment above FMT_REDUCE_INT_INSTANTIATIONS definition describing
why a developer would use it.

* Move FMT_REDUCE_INT_INSTANTIATIONS to format.h

Co-authored-by: Khalil Estell <kammce@google.com>
---
 include/fmt/format.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/include/fmt/format.h b/include/fmt/format.h
index 6548f716..3a9cf9b3 100644
--- a/include/fmt/format.h
+++ b/include/fmt/format.h
@@ -163,6 +163,14 @@ FMT_END_NAMESPACE
 #  define FMT_USE_LONG_DOUBLE 1
 #endif
 
+// Defining FMT_REDUCE_INT_INSTANTIATIONS to 1, will reduce the number of
+// int_writer template instances to just one by only using the largest integer
+// type. This results in a reduction in binary size but will cause a decrease in
+// integer formatting performance.
+#if !defined(FMT_REDUCE_INT_INSTANTIATIONS)
+#  define FMT_REDUCE_INT_INSTANTIATIONS 0
+#endif
+
 // __builtin_clz is broken in clang with Microsoft CodeGen:
 // https://github.com/fmtlib/fmt/issues/519
 #if (FMT_GCC_VERSION || FMT_HAS_BUILTIN(__builtin_clz)) && !FMT_MSC_VER
@@ -738,12 +746,19 @@ FMT_CONSTEXPR bool is_supported_floating_point(T) {
          (std::is_same<T, long double>::value && FMT_USE_LONG_DOUBLE);
 }
 
+#if FMT_REDUCE_INT_INSTANTIATIONS
+// Pick the largest integer container to represent all values of T.
+template <typename T>
+using uint32_or_64_or_128_t =
+    conditional_t<sizeof(uint128_t) < sizeof(uint64_t), uint64_t, uint128_t>;
+#else
 // Smallest of uint32_t, uint64_t, uint128_t that is large enough to
 // represent all values of T.
 template <typename T>
 using uint32_or_64_or_128_t = conditional_t<
     std::numeric_limits<T>::digits <= 32, uint32_t,
     conditional_t<std::numeric_limits<T>::digits <= 64, uint64_t, uint128_t>>;
+#endif
 
 // Static data is placed in this class template for the header-only config.
 template <typename T = void> struct FMT_EXTERN_TEMPLATE_API basic_data {