Date: Tue, 16 Dec 2014 12:47:16 -0700 From: Warner Losh <imp@bsdimp.com> To: Dimitry Andric <dim@freebsd.org> Cc: freebsd-toolchain@freebsd.org Subject: Re: Migration to dynamic libs for llvm and clang Message-ID: <098F3C6B-6DB4-472F-9CA0-D12BE5E37A19@bsdimp.com> In-Reply-To: <6D28CE1D-B94D-4E5E-B2DC-96C47E6C63BF@FreeBSD.org> References: <CAPyFy2DeLiFAW_yS14r1n89r92MFG1sbX88rNgaJshwH9-%2BkQg@mail.gmail.com> <41F09A1C-01D6-42C9-B495-244DFC2B0364@FreeBSD.org> <D359161D-B14C-4F19-8F0D-57FE530D0AF4@FreeBSD.org> <74C51AC7-B7ED-4EBC-8506-1554C7CA31FF@FreeBSD.org> <CAPyFy2BSU%2B8-TWh53z_FT-z2NCsCDU6=%2Bi_-OH6MJTxo2dqhpw@mail.gmail.com> <15A17E31-22CA-4680-869B-3B7AC1E49741@bsdimp.com> <6D28CE1D-B94D-4E5E-B2DC-96C47E6C63BF@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--Apple-Mail=_A5ED19A7-AC5E-440E-8907-8DE594C8A531 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 > On Dec 16, 2014, at 12:01 PM, Dimitry Andric <dim@freebsd.org> wrote: >=20 > On 16 Dec 2014, at 18:54, Warner Losh <imp@bsdimp.com> wrote: >>=20 >>> On Dec 16, 2014, at 10:44 AM, Ed Maste <emaste@freebsd.org> wrote: >>>=20 >>> Fair enough, I'd definitely like to see fewer build-time knobs over >>> time, not more. >>=20 >> Until we stop using build-time knobs to control what=E2=80=99s in the = final image >> as a poor man=E2=80=99s packaging scheme, I expect the number of = knobs to >> continue to grow. >=20 > How does a packaging scheme solve the problem of not compiling in > dependencies, or linking everything static? You cannot pre-build all > possible combinations of selectable options. >=20 > As for knobs that just say "build foo" or "don't build bar", those = would > indeed be fine for a packaging system, as long as packages aren't too > dependent on each other. Right now we mix build options for building things or not (e.g. = WITH_SENDMAIL) with build options for things like this (WITH_SHARED_TOOLCHAIN). The = number of the former is increasing all the time (with a big increase when = Ngie=E2=80=99s work hits the tree). So in many senses, it is an orthogonal issue. My comment was more to Ed=E2=80=99s notion that these numbers will be dropping any time = soon. packing the base is actually a hard problem because the phrase =E2=80=9Cas= long as packages aren=E2=80=99t dependent on each other=E2=80=9D turns out to = not be the case as much as we might like. The base is fairly interdependent of all that = since we have it as one big ball of stuff right now. Most of the issues revolve = around dependence on the libraries and such. With patience and diligence, I = think we can package the base, but it isn=E2=80=99t going to be a trivial slam = dunk if it is going to be useful. Warner --Apple-Mail=_A5ED19A7-AC5E-440E-8907-8DE594C8A531 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - https://gpgtools.org iQIcBAEBCgAGBQJUkIxFAAoJEGwc0Sh9sBEAxzQP/2o98z/7Y9Tlc3ZrxHBxHU+X ij1svutS4QQV210j3JYAvKEgWBQambnAg1rO2uFhg48zbZd+Zf8SFe/l0+Tf7oGm aCE2/bONVJrGesR10DkHWXfx50Z9dqIfzapZQJ+z6//NHMO7bSYhFd1XFUQB9xIG WQfpJAX3BPu74reIsfHmT4R32BRlVdSY+4LX8AjLqtDSCTDib96pnDj6EwjFl79e NoA39WXG2C/eg6OZQyDX0q29nVqLpZNEh2o+WL+H1cOU48KE0ezotAtr4nt/Kwrl 2U9s6WP+QZkXv9Bd6XKz4h4DEdlz3pVmlg2F4qbz0pAUjPbGaltRQl2tFBFwcqcQ 9Jxapy27tDV0tIX0J29ZrTLT6Hn41FOeXDH9OeRa0ZIiElXMXUaRssm6yLr4B8II KpCUJ1cXxefNfasfyyZKxudG4efIvDpsyhkqwY1sWHnY9xi/YcYGZZ2HJj0/aJxA y25preqXFwUQr0/yYEg98WIGMsWEQPB8QY8XA7tNGNJWWzsN8gWxRdaYSd0zpMa7 vXPNsyNmPx9DntHFhT8/iYRvFr9jmg1y1aGffxClkx7yQfODPuweWRKlkWK/If/u g+tlBlKcl06yXf1SSeXtVzkVQ+SAtjYpQGk6nTryT6WGQOj3KNSfZo+lxNQ5Wqow MhS8C8Pz0cJFhL6FDUM7 =wTKG -----END PGP SIGNATURE----- --Apple-Mail=_A5ED19A7-AC5E-440E-8907-8DE594C8A531--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?098F3C6B-6DB4-472F-9CA0-D12BE5E37A19>