Date: Tue, 16 Feb 2016 16:31:06 -0800 From: Bryan Drewery <bdrewery@FreeBSD.org> To: David Wolfskill <david@catwhisker.org>, current@freebsd.org Subject: Re: Possible race condition noted r295622 -> r295654 Message-ID: <56C3BF4A.4040104@FreeBSD.org> In-Reply-To: <20160216150249.GT1178@albert.catwhisker.org> References: <20160216150249.GT1178@albert.catwhisker.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2/16/16 7:02 AM, David Wolfskill wrote:
> Given some of the reports of problems building head recently, my initial
> attempts this morning (upgrading head/amd64) in place from r295622 to
> r295654) were "clean" builds.
>
> My "build machine" reported no trouble (using -j16).
>
> On my laptop, also using -j16, the process terminated prematurely, with
> whines about inability to find libopythagoras.
>
> In a half-hearted attempt, I then immediately attempted an "unclean"
> build ("make -j 16 --DNO_CLEAN buildworld"); unsurprisingly, that
> failed.
>
> On a bit of a whim, I then tried "make -j14 buildworld" -- that
> succeeeded.
>
> I've placed a typescript of all of those efforts in
> <http://www.catwhisker.org/~david/FreeBSD/head/>; -- ref "bw.txt" (or
> "bw.txt.gz"), in case that's of interest or use.
>
r295683 should fix it. There are likely more regressions after my recent
change. Please CC me on any more you run into.
--
Regards,
Bryan Drewery
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?56C3BF4A.4040104>