)]}'
{
  "commit": "5831832cd81027cc8588138488bccb5854e7261d",
  "tree": "20f41f0084c1369c85c87d70a4eac05c20a45a6c",
  "parents": [
    "5aff1006c1d94600a078c1cae9aea729dc69febd"
  ],
  "author": {
    "name": "Michael Matloob",
    "email": "matloob@golang.org",
    "time": "Fri Jul 11 16:36:51 2025 -0400"
  },
  "committer": {
    "name": "Michael Matloob",
    "email": "matloob@google.com",
    "time": "Wed Jul 23 08:02:37 2025 -0700"
  },
  "message": "sweet/benchmarks/go-build: build runtime before running benchmark\n\nSo we don\u0027t consider time building runtime when comparing go build time.\nInstead of using go build -a to rebuild everything, we\u0027ll use a new\ntemporary build cache directory so that nothing is cached. Then we\u0027ll\nbuild the runtime using that cache directory to make sure that it is\ncached. Then the benchmark run build, without -a will build everything\nexcept for the runtime and its dependencies, which have already been\ncached, and that time will be used in the benchmark result.\n\nChange-Id: I2753fc8d8438be1d437f7c65077ea22760a34b54\nReviewed-on: https://go-review.googlesource.com/c/benchmarks/+/688915\nReviewed-by: Michael Matloob \u003cmatloob@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Michael Knyszek \u003cmknyszek@google.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "c3c9a803b37cb946b779fafa547a875a0fd519ba",
      "old_mode": 33188,
      "old_path": "sweet/benchmarks/go-build/main.go",
      "new_id": "593470699f95e9fe273d21be66e93f051697f5dc",
      "new_mode": 33188,
      "new_path": "sweet/benchmarks/go-build/main.go"
    }
  ]
}