| commit | 0bb7f258ab0f63f354a3eedb58c950ae9ba5593d | |
| author | Catherine Gasnier <catg@fb.com> | |
| Tue, 12 Apr 2022 10:46:39 +0000 (12 03:46 -0700) | ||
| committer | Facebook GitHub Bot <facebook-github-bot@users.noreply.github.com> | |
| Tue, 12 Apr 2022 10:46:39 +0000 (12 03:46 -0700) | ||
| tree | ce7e82e524bef8ab12978afed39236ac49cfd531 | treesnapshot (tar.gz zip) |
| parent | 7ab5c516440c38fd63a3f4d8112f3fd88c7a8be2 | commitdiff |
Telemetry: remove is_lazy_incremental
Summary:
I don't trust how is_lazy_incremental is handled. It's a global reference that sometimes gets set with a setter. Looking at code around that reference in HackEventLogger, it seems non-trivial to get it right, and impossible to convince oneself it is correct.
Besides, for some events like recheck_end event, for example, it's completely meaningless because a "recheck" is a loop with multiple checks per iterations, where each check can either be lazy or full. We now have a check_kind field which is properly set with a with_check_kind helper, so is_lazy_incremental is redundant.
Reviewed By: ljw1004
Differential Revision: D35547976
fbshipit-source-id: a9c0f23e19f1ed9029ddc36c7c93fc0cdc83952e
Summary:
I don't trust how is_lazy_incremental is handled. It's a global reference that sometimes gets set with a setter. Looking at code around that reference in HackEventLogger, it seems non-trivial to get it right, and impossible to convince oneself it is correct.
Besides, for some events like recheck_end event, for example, it's completely meaningless because a "recheck" is a loop with multiple checks per iterations, where each check can either be lazy or full. We now have a check_kind field which is properly set with a with_check_kind helper, so is_lazy_incremental is redundant.
Reviewed By: ljw1004
Differential Revision: D35547976
fbshipit-source-id: a9c0f23e19f1ed9029ddc36c7c93fc0cdc83952e
| hphp/hack/src/server/serverTypeCheck.ml | diffblobblamehistory | |
| hphp/hack/src/stubs/logging/hackEventLogger.ml | diffblobblamehistory |