pkgsrc-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Failure building firefox-65.0



This is probably true. I tried to ktrace the two cargo instances when it was wedged, did not get any response. 


On Sat, 2 Feb 2019, 07:49 Tobias Nygren <tnn%netbsd.org@localhost wrote:
On Fri, 1 Feb 2019 23:51:17 +0100
bch <brad.harder%gmail.com@localhost> wrote:

> On Fri, Feb 1, 2019 at 23:17 Chavdar Ivanov <ci4ic4%gmail.com@localhost> wrote:
>
> > Eventually completed and working fine. First attempt had cargo hung, I
> > waited for parhaps an hour, there was no activity, so I interrupted
> > and restarted make (without cleaning, I've had that previously).
> >
>
> This build (because of rust) is so crazy. It?s nearly at the place where it
> alone needs a dedicated build/storage infrastructure.
>
> Is there already (or interest in) a build for we on nbsd -current and
> pkgsrc -current? If there?s something available, I?d love to know because
> it?s at least a multi-day process for me if I?m lucky.

I run pkgsrc on both NetBSD and Linux (on work laptop). The rust parts
of the Firefox build go significantly faster on Linux and don't get
stuck so I think there is some BSD specific problem with rust.

A hint might be to look in top(1) when it gets wedged. The wchan column
of rust processes sometimes shows a garbage state for me. It's
easier to reproduce while building rust itself than firefox.

-Tobias


Home | Main Index | Thread Index | Old Index