Re: [PATCH] travis-ci: run previously failed tests first, then slowest to fastest

From: Jeff King <peff@peff.net>
Date: Jan 20, 2016, 01:56 UTC
Message-ID: <20160120015602.GC24541@sigill.intra.peff.net>
In-Reply-To: <xmqqfuxt6n00.fsf@gitster.mtv.corp.google.com>

On Tue, Jan 19, 2016 at 05:46:55PM -0800, Junio C Hamano wrote:

Show 7 quoted lines

> > "Travis CI VMs run on 1.5 virtual cores."
> 
> Yup, that 1.5 was already mentioned in the earlier thread, but many
> tests are mostly I/O bound, so 1.5 (or 2 for that matter) does not
> mean we should not go higher than -j2 or -j3.  What I meant was that
> the 3 comes from the old discussion "let's be nice to those who
> offer this to us for free".

I am very appreciative that we can use Travis for free, but I doubt they care much one way or the other how we parallelize. Everything is sandboxed enough that we should not be able to cause problems for them or other customers. It's all CPU seconds to them (or should be, anyway).

The thing that _would_ probably bother them is throwing too many builds at it (right now we are building the integration branches; it would be useful information to build individual topics, too, but that would increase the number of CPU seconds we ask them for).

-Peff