What's going on with exp() and --math-mode=fast?

General Usage
27

Hi, I don’t know if this is related or fixed already. I tracked down an issue (https://github.com/JuliaRobotics/IncrementalInference.jl/issues/1382) in julia 1.7rc1 to a function that uses @fastmath in https://github.com/JuliaRobotics/KernelDensityEstimate.jl .

for julia 1.7.0-rc1

julia> versioninfo()
Julia Version 1.7.0-rc1
Commit 9eade6195e (2021-09-12 06:45 UTC)
Platform Info:
  OS: Linux (x86_64-pc-linux-gnu)
  CPU: Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-12.0.1 (ORCJIT, skylake)
Environment:
  JULIA_EDITOR = code
  JULIA_NUM_THREADS = 4

julia> foo(x) = @fastmath exp(-0.5*x)
foo (generic function with 1 method)

julia> foo(2835.0)
-7.888656068042933

julia> exp(-0.5*2835.0)
0.0

for julia 1.6.3

julia> versioninfo()
Julia Version 1.6.3
Commit ae8452a9e0 (2021-09-23 17:34 UTC)
Platform Info:
  OS: Linux (x86_64-pc-linux-gnu)
  CPU: Intel(R) Core(TM) i5-7300HQ CPU @ 2.50GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-11.0.1 (ORCJIT, skylake)

julia> foo(x) = @fastmath exp(-0.5*x)
foo (generic function with 1 method)

julia> foo(2835.0)
0.0

julia> exp(-0.5*2835.0)
0.0
1 Like