René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Quicken everything","articleBody":"...as discussed in https://github.com/faster-cpython/ideas/issues/462.\r\n\r\nBasically, we can simplify the quickening process by quickening code objects as they are created. This works by ditching our per-code-object warmup counter, and instead using our adaptive backoff counters for more-granular \"warming\".\r\n\r\nI have a branch that implements this change, resulting in consistent a ~1% performance improvement. I imagine that this comes from several places:\r\n- A lower specialization threshold. I tried a wide range of initial adaptive counter values (ranging from 0 to 4095), and a value of 1 worked best. With this value, adaptive instructions will specialize the second time they are run.\r\n- A simpler interpreter loop. `RESUME_QUICK` and `JUMP_BACKWARD_QUICK` are no longer needed.\r\n- Superinstructions everywhere. All code objects have superinstructions now.\r\n\r\nNext steps will be simplifying the unmarshal / code creation process as part of this same work.\n\n\u003c!-- gh-pr-number: gh-98687 --\u003e\n* PR: gh-98687\n\u003c!-- /gh-pr-number --\u003e\n\n\n\u003c!-- gh-pr-number: gh-99106 --\u003e\n* PR: gh-99106\n\u003c!-- /gh-pr-number --\u003e\n\n\n\u003c!-- gh-pr-number: gh-99182 --\u003e\n* PR: gh-99182\n\u003c!-- /gh-pr-number --\u003e\n\n\n\u003c!-- gh-pr-number: gh-99399 --\u003e\n* PR: gh-99399\n\u003c!-- /gh-pr-number --\u003e\n","author":{"url":"https://github.com/brandtbucher","@type":"Person","name":"brandtbucher"},"datePublished":"2022-10-25T23:48:33.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":1},"url":"https://github.com/98686/cpython/issues/98686"}