I have recently been reading the SBCL User Manual and started wondering about the title question. Obviously some lisps, for example clojure, ban all side effects so they can easily parallelize the code. Common Lisp allows side effects and so I was wondering if the fact a given function is 'dirty' or 'clean' affects it's compilation.
For example in the CMUCL compiler manual let optimizations show how, in many casesm the use of 'let' to bind a new variable will be more efficient than modifying with 'setq'. I guess I'm asking if something similar is done for function calls.
I have read the relevant sections of the sbcl manual and poured through the question on stackoverflow but could not find an answer to this.
At 25000 values, Common Lisp is almost 1.8 times as fast as the C version, and the compilation time is 65% of the total evaluation time. In Figure 2, for programs of depth 8, Common Lisp passes C at between 5000 and 6000 values, and the compilation time is 16.2 seconds.
To quit SBCL, type (quit) .
short: Not faster. Sometimes actually slower.
long:
According to Stas Boukarev from SBCL-devel,
SBCL doesn't even know that a function has no side effects, so, no. Besides, most of the time having side effects is the most optimal way.
I am aware of the fact that functions such as nreverse
, which are destructive, tend to be faster than nondestructive functions (in this case reverse
is the nondestructive version). They also come with many setbacks. As Peter Siebel put it:
Each recycling function is a loaded gun pointed footward.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With