Date: Wed, 20 Feb 2019 18:14:24 +1100 From: Kubilay Kocak <koobs@FreeBSD.org> To: yuri@freebsd.org, ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org Subject: Re: svn commit: r493408 - head/graphics/oidn Message-ID: <cfb359aa-2857-da60-b2f0-a2b55e7aeb15@FreeBSD.org> In-Reply-To: <2d467c66-b275-24eb-57f5-25ac1fcbd964@freebsd.org> References: <201902200622.x1K6MAqG093448@repo.freebsd.org> <daa1b9b2-9df8-7ca0-4ecf-9bc3e529e7ef@FreeBSD.org> <2d467c66-b275-24eb-57f5-25ac1fcbd964@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 20/02/2019 6:10 pm, Yuri wrote: > On 2/19/19 10:44 PM, Kubilay Kocak wrote: >> On 20/02/2019 5:22 pm, Yuri Victorovich wrote: >>> +BROKEN_i386= Intel(R) MKL-DNN supports 64 bit platforms only >>> (oidn contains a custom version of mkl-dnn, see math/mkl-dnn) >>> + >> >> Could this then be ONLY_FOR_ARCHS or NOT_FOR_ARCHS ? >> > > I'm not sure if it breaks on other architectures. There are people who > regularly build on other architectures and they didn't complain about > math/mkl-dnn which has the same architecture specificity. Fair. Well considered. Such is the BROKEN <-> *_FOR_ARCHS trade-off.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?cfb359aa-2857-da60-b2f0-a2b55e7aeb15>