master-chromium-pfq of R62 picks R61 afdo profiles |
||||
Issue description00:08:22: INFO: Found latest AFDO info file gs://chromeos-prebuilt/afdo-job/llvm/latest-chromeos-chrome-amd64-62.afdo but it is too old ... /b/c/cbuild/repository/chroot/tmp/latest-chromeos-chrome-amd64-61.afdo 00:08:22: INFO: Found previous x86 AFDO profile chromeos-chrome-amd64-61.0.3163.60_rc-r1.afdo IMHO, it should FAIL rather than picking up something from R61. The logic should be: 1. when latest-* is not found: try version - 1. 2. when latest-* is too old: break
,
Sep 28 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromiumos/chromite/+/659844701de192fa02a38f1645795ad1230ccd8f commit 659844701de192fa02a38f1645795ad1230ccd8f Author: Ting-Yuan Huang <laszio@chromium.org> Date: Thu Sep 28 04:12:55 2017 afdo: fix the logic determining profile expiration When there has been no profile generated since last branch point, the waterfall will pick the latest profile from the last branch. There is a corner case undesirable in the original logic: when a bug creeps in after the branch point and before the first profile on master is generated, the waterfall will keep picking up profiles from the latest branch, as long as there are new profiles available, in stead of reporting error. BUG= chromium:758318 TEST=cros tryjob master-chromium-pfq Change-Id: I32ec0f2263b31d91ccce802a1a148ca7352782ba Reviewed-on: https://chromium-review.googlesource.com/688156 Commit-Ready: Ting-Yuan Huang <laszio@chromium.org> Tested-by: Ting-Yuan Huang <laszio@chromium.org> Reviewed-by: Ting-Yuan Huang <laszio@chromium.org> [modify] https://crrev.com/659844701de192fa02a38f1645795ad1230ccd8f/cbuildbot/afdo.py
,
Sep 28 2017
,
Jan 22 2018
,
Jan 23 2018
|
||||
►
Sign in to add a comment |
||||
Comment 1 by llozano@chromium.org
, Aug 23 2017Status: Assigned (was: Untriaged)