miniblog.
← Back to all posts
Wilfred Hughes
Jan 11, 2014 at 23:36
LLVM IR rejects 1.3 as a floating-point constant, since it can't be represented exactly in binary. Smart way to avoid confusion.