[PATCH] D49878: [Fixed Point Arithmetic] Fixed Point log2(), log10(), and pow()

[Leonard Chan via Phabricator via llvm-commits]

leonardchan created this revision.
leonardchan added reviewers: phosek, mcgrathr.
leonardchan added a project: clang.
Herald added subscribers: Sanitizers, llvm-commits, mgorny.

This patch is not actually up for review and meant to hold the remaining work done after https://reviews.llvm.org/D49876 but haven't submitted for review yet.

This patch contains:

- The implementation for `log2()`, `log10()`, and `pow10()`. Each of these functions takes a long _Accum and returns a value with 31 bits of precision and the value returned is less than one unit of precision away from the actual value of log() and pow() (that is error < 1/2^31).


Repository:
  rCRT Compiler Runtime

https://reviews.llvm.org/D49878

Files:
  cmake/builtin-config-ix.cmake
  lib/builtins/CMakeLists.txt
  lib/builtins/lognfx.c
  lib/builtins/pow10fx.c
  test/builtins/Unit/lognfx.c
  test/builtins/Unit/pow10fx.c

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D49878.157586.patch
Type: text/x-patch
Size: 23767 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180726/5ff0b0bc/attachment.bin>