diff options
| author | Cherry Mui <cherryyz@google.com> | 2026-03-25 09:41:27 -0400 |
|---|---|---|
| committer | Cherry Mui <cherryyz@google.com> | 2026-03-25 08:34:09 -0700 |
| commit | fa238516b782bd1f233e85b719b7ab90889a5634 (patch) | |
| tree | bbbf3b3f0cdfc460d060fcdf251531025bf7ac36 /src/database/sql | |
| parent | b8dea877130ee6b80df1c468952ec5d13de9c16a (diff) | |
| download | go-fa238516b782bd1f233e85b719b7ab90889a5634.tar.xz | |
cmd/link: don't run TestExtLinkCmdlineDeterminism in parallel with other tests
Based on the investigation of #78180, it appears that for
different runs the go command picks up different object from the
build cache for the runtime/cgo package. One possible situation
could be that as the test runs in parallel with other tests, some
other test builds runtime/cgo at same time, and somehow that build
is not deterministic, causes different objects of runtime/cgo
being written to the build cache. Let's see if running the test
not in parallel helps. The test process may still run in parallel
with other processes, which could also build runtime/cgo and
interfere, so this may be just best effort. If the build of
runtime/cgo is deterministic, this shouldn't be necessary.
For #78180.
Change-Id: If078f141febd33f063fe2d564c4150223a9784ef
Reviewed-on: https://go-review.googlesource.com/c/go/+/759122
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Diffstat (limited to 'src/database/sql')
0 files changed, 0 insertions, 0 deletions