On 2015/06/10 17:00:13, Michael Achenbach wrote:
On 2015/06/10 14:24:33, arv wrote:
> I think the underlying problem is here:
>
> https://github.com/tc39/test262/issues/295
>
> The "filenames" in the includes directive ended up containing an \r
which
> explains why the files could not be found
Yea, that sounds like it. And the flakiness can maybe be explained with
the
tester slaves changing between the runs. There might be some differences
from
bot to bot how to handle this. This CL here, however, is still useful as
we
otherwise aren't even able to cleanly revert a test262 upgrade.
There is more to it. Some bots have a yaml python package but not all do. If
they don't we use something called monkeyYaml which is part of the test262
harness. I've run into issues before where these two yaml implementations
diverged and caused some bots to pass and some to fail.
Would we need an upstream fix for this?
I don't think this applies to the upstream test harness.
https://codereview.chromium.org/1173753004/
--
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to v8-dev+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
