Date: Sun, 27 Oct 2019 21:52:50 -0500 From: Jason Bacon <bacon4000@gmail.com> To: Yuri <yuri@rawbw.com>, Tobias Kortkamp <tobik@freebsd.org> Cc: Raphael Kubo da Costa <rakuco@freebsd.org>, ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org Subject: Re: svn commit: r515803 - head/devel/bazel Message-ID: <1d688197-0ca7-025d-bdcd-4d23631182b2@gmail.com> In-Reply-To: <778754a2-2aa5-831a-7441-a9a59c25b512@rawbw.com> References: <201910271742.x9RHgvQP017264@repo.freebsd.org> <09d56af7-96f5-dcfd-d70c-d20d54e97ea2@rawbw.com> <20191027222033.GA81623@urd.tobik.me> <778754a2-2aa5-831a-7441-a9a59c25b512@rawbw.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2019-10-27 17:23, Yuri wrote: > On 2019-10-27 15:20, Tobias Kortkamp wrote: >> No, Tensorflow has been failing to build for some time now and was >> probably broken during the bazel 1.0.0 update in r514543. Definitely >> before r515803 [1]. > > > This means that several people (>1) commit updates without testing them= =2E > > > Yuri > > I think we do need to shore up our efforts to avoid this sort of=20 breakage.=A0 If a package like tensorflow can't be installed/upgraded=20 reliably, this could delay someone's research and scare them away from=20 using FreeBSD for scientific computing. If we avoid issues like this, FreeBSD ports can be advertised as a big=20 advantage, since many scientific packages lack a reliable and performant = installation method on other platforms. TensorFlow isn't the greatest poster child for us, since many users need = CUDA to get good performance out of it, but that's very rare and the=20 principal here applies to most other scientific software. Cheers, =A0=A0=A0 JB --=20 Earth is a beta site.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1d688197-0ca7-025d-bdcd-4d23631182b2>