To be fair for the C version: The linked code [1] doesn't really look like "decades of optimized C" to me. More like a straightforward implementation of this, w

To be fair for the C version: The linked code [1] doesn't really look like "decades of optimized C" to me. More like a straightforward implementation of this, which had been likely had not seen many updates for decades. It could probably gain performance by playing around with input buffer sizes and trying to make use of vectorized instructions (the compiler might try to auto-vectorize the lops, but there is no guarantee). [1] https://opensource.apple.com/source/text_cmds/text_cmds-68/wc/wc.c.auto.html https://opensource.apple.com/source/text_cmds/text_cmds-68/w...