{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreihtu6nh25ogcrwefckzbjtiegmh5s3gpocfqww2l2qb5nrnuwmie4",
    "uri": "at://did:plc:b3tz6srl4ochk2wxn6dv6xpy/app.bsky.feed.post/3mifee6gclzk2"
  },
  "path": "/Articles/1064830/",
  "publishedAt": "2026-03-31T15:40:56.000Z",
  "site": "https://lwn.net",
  "tags": [
    "a memory-management patch set",
    "proposed",
    "Sashiko",
    "recently released LLM-based kernel patch review system"
  ],
  "textContent": "Discussion of \na memory-management patch set intended to clean up a helper function for handling huge pages spiraled into something else entirely after it was posted on March 19. Memory-management maintainer Andrew Morton \nproposed making changes to the subsystem's review process, to require patch authors to respond to feedback from Sashiko, the \nrecently released LLM-based kernel patch review system. Other sub-maintainers, particularly Lorenzo Stoakes, objected. The resulting discussion about how and when to adopt Sashiko is potentially relevant to many other parts of the kernel.",
  "title": "[$] The role of LLMs in patch review"
}