[llvm] r229841 - Reverting r229831 due to multiple ARM/PPC/MIPS build-bot failures.

[Renato Golin]

On 13 May 2015 at 11:39, Kuperstein, Michael M
<michael.m.kuperstein at intel.com> wrote:
> http://lab.llvm.org:8011/builders/clang-cmake-thumbv7-a15-full-sh/builds/948/steps/build%20stage%201/logs/stdio
>
> command timed out: 1200 seconds without output, attempting to kill
> process killed by signal 9
> program finished with exit code -1
> elapsedTime=3712.616572

Damn it, I can't reproduce it. On a Chromebook 2, it builds ok, tests
ok, selfhosts ok, tests selfhosted ok.

The difference from the buildbot is that they have 2 cores and 2GB or
RAM (+2GB swap), while the Chrome2s have 4 cores and 4GB or RAM (+4GB
swap), which may exaplain why they don't run out of memory or timeout
(supposedly swapping).

I don't want to take the buildbot off the rota, so I'll try ulimit to
reduce the amount of memory per process... Or see if I can dig a
Chromebook 1.

Did you manage to reproduce on x86?

cheers,
--renato