On 2012-10-01 4:03 PM, Kyle Huey wrote:
On Mon, Oct 1, 2012 at 1:01 PM, Ehsan Akhgari <ehsan.akhg...@gmail.com
<mailto:ehsan.akhg...@gmail.com>> wrote:

    On 2012-09-30 2:47 AM, Justin Lebar wrote:

                Unfortunately, due to
                https://bugzilla.mozilla.org/__show_bug.cgi?id=767501
                <https://bugzilla.mozilla.org/show_bug.cgi?id=767501>,
                the only way to get b2g builds on try is to use '-p all'.



            Sounds like that should be a P1 for reducing load then.


        Much-to-most of this "B2G-only" code compiles on other
        platforms, too.
           Or at least, that is true of the B2G patches I've reviewed.  We
        shouldn't exaggerate the effect of this without some data.
          (Note that
        basically any code we want to write unit tests for at the moment
        must
        compile on more than B2G, so it's not entirely unintentional that so
        many patches touch other platforms.)


    If all they care about is whether their code compiles, they should
    compile locally on a different platform.  Try is not meant to
    replace local builds, really, and should not be used as merely a
    compiler.


You expect all of our developers to be able to compile the at least 6[1]
separate platforms that we have?

No, but I expect them to be able to compile on one non-b2g platform. I push stuff that I have only tested to build on one platform all the time without making them go through try, and I'm saving resources doing that even though I occasionally break the tree on a platform I have not built on. I refer you to jlebar's analysis in the other thread we had on this a while ago.

Ehsan

_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform

Reply via email to