)]}'
{
  "commit": "f64daeb7efebf3b9ecd00eb04d279ea8411c5c24",
  "tree": "a225259034b92b1ee84ba92b524697a0c5e7e4c4",
  "parents": [
    "06bd998e21a22f7ba68e480ce5cdf43102888534"
  ],
  "author": {
    "name": "Alan Donovan",
    "email": "adonovan@google.com",
    "time": "Thu Oct 08 17:41:49 2026 -0400"
  },
  "committer": {
    "name": "Alan Donovan",
    "email": "adonovan@google.com",
    "time": "Tue Oct 20 17:37:11 2026 -0700"
  },
  "message": "gopls/internal/filecache: reduce I/O parallelism\n\nThis CL reduces the parallelism for writes to 8 (was 128).\nThis reduced time of a cold \"gopls check\" of k8s kubelet:\n- wall 22.2s -\u003e 20.0s (n\u003d3)\n- sys  46.9s -\u003e 42.6s.\n\nOn macOS, Set/Get account for 20% of CPU, mostly in open(2).\n(Both sys and wall time in open were 4-5x higher before.)\n\nSplitting the Set and Get operations into separate\nsemaphores was evaluated but not found profitable.\n\nChange-Id: Icae8e626f549df2fd5dcfedacf58211b2ad52850\nReviewed-on: https://go-review.googlesource.com/c/tools/+/847645\nReviewed-by: Peter Weinberger \u003cpjw@google.com\u003e\nReviewed-by: Alex Putman \u003caputman@golang.org\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "44db48e1533df253f1f409a597a4bb12a586acf2",
      "old_mode": 33188,
      "old_path": "gopls/internal/filecache/filecache.go",
      "new_id": "0c396983a15a3e251f2dee6bdfdc8873978ede57",
      "new_mode": 33188,
      "new_path": "gopls/internal/filecache/filecache.go"
    },
    {
      "type": "modify",
      "old_id": "06fa96787c2f5b4732ce18c48e22a49504c32462",
      "old_mode": 33188,
      "old_path": "gopls/internal/filecache/filecache_test.go",
      "new_id": "ccc211a40612d06ef0d3a9a44f1984e924538887",
      "new_mode": 33188,
      "new_path": "gopls/internal/filecache/filecache_test.go"
    }
  ]
}