r283680 - [CUDA] Support <complex> and std::min/max on the device.

[Reid Kleckner via cfe-commits]

On Tue, Oct 11, 2016 at 9:34 AM, Justin Lebar <jlebar at google.com> wrote:

> > Can we figure out why having an incremental stage1 is problematic? I
> want to keep it that way to speed up the bot.
>
> The problem is that the cmake install rule that we use does not remove
> extra files that happen to be present in the install directory.  So if
> you put something bad in there, as I did, it's never coming out.
>
> I spent a few hours trying to figure out how to do this yesterday to no
> avail.
>
> But it would be nice not to block this change on fixing our cmake
> rules, especially if the relevant expertise is hard to come by.


Sure. I don't think the "install" step is expensive. We should be able to
fix this problem going forward by changing our bot config to clean out the
installation directory before the install step.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20161011/64b6ce11/attachment.html>