From mboxrd@z Thu Jan 1 00:00:00 1970 From: Adam Van Ymeren Subject: =?UTF-8?Q?Re=3A_Roadmap_for_Guix=C2=A01=2E0?= Date: Tue, 04 Sep 2018 19:09:36 -0400 Message-ID: <97A44D46-4F09-4C76-9533-A77240C18981@vany.ca> References: <878t5udq9u.fsf@gnu.org> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="----6CQ62YXE3T38D9Q66QB8XOGPKAC2JZ" Content-Transfer-Encoding: 7bit Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:54961) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fxKSD-0005e4-6C for guix-devel@gnu.org; Tue, 04 Sep 2018 19:09:50 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fxKS6-0004EC-2l for guix-devel@gnu.org; Tue, 04 Sep 2018 19:09:47 -0400 In-Reply-To: List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: =?ISO-8859-1?Q?Ludovic_Court=E8s?= Cc: "guix-devel@gnu.org" ------6CQ62YXE3T38D9Q66QB8XOGPKAC2JZ Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Kinda late to the party here but I think a goal for 1=2E0 should be to ensu= re every single package builds on x86_64 and/or i686 and that most substit= utes are available at the time of release=2E Having guix claim to have packages which then fail to build can leave a po= or first impression=2E It's fine for alpha/beta but I think it will be rea= lly important for 1=2E0 to feel as robust as possible=2E I'd even say that any packages which don't build for 1=2E0 should be delet= ed or at least hidden=2E Maybe of channels happen we could use a stable vs = unstable channels=2E That's my 2 cents on 1=2E0 at least=2E On September 4, 2018 6:55:57 PM EDT, fis trivial wrote: > >Ludovic Court=C3=A8s writes: > >> Hello Guix! >> >> I=E2=80=99ve pushed to guix/maintenance=2Egit a list of things that IMO= we >should >> do or might want to do for 1=2E0, with the understanding that 1=2E0 >should >> happen in 2018 (or early 2019 at the latest!)=2E :-) >> >> =20 >https://git=2Esavannah=2Egnu=2Eorg/cgit/guix/maintenance=2Egit/tree/doc/1= =2E0=2Eorg >> >> The list focuses on =E2=80=9Cbig item=E2=80=9D features and tasks, omit= ting routine >bug >> fixes and improvements=2E Some of these items don=E2=80=99t require a = lot of >work >> or expertise though, so hopefully there=E2=80=99s enough on everyone=E2= =80=99s plate=2E >> >> Feel free to comment, volunteer, add items (but not too many!), >remove >> items, promote items, etc=2E Committers should feel free to edit the >file >> directly in maintenance=2Egit, especially to mark things as done=2E ;-= ) >> >> Copy of the file attached below=2E >> >> Ludo=E2=80=99=2E >> >> PS: I=E2=80=99m starting the discussion but will go AFK soon after send= ing >this >> message=2E :-) > >Hi Ludovic, > >For near future version of GUIX, maybe 1=2E1, would it be possible to: > > Upgrade to latest GCC as base compiler > > Use gold linker as default > >Based on experience, newer GCC and gold linker could reduce compilation >time >and memory requirement dramatically=2E It would be beneficial to both >build >farm which have heavy load and laptop users who have very limited >memory=2E > >Thanks=2E >--=20 >Jiaming ------6CQ62YXE3T38D9Q66QB8XOGPKAC2JZ Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable Kinda late to the party here but I think a goal fo= r 1=2E0 should be to ensure every single package builds on x86_64 and/or i= 686 and that most substitutes are available at the time of release=2E

Having guix claim to have packages which then fail to build can leave a po= or first impression=2E It's fine for alpha/beta but I think it will be= really important for 1=2E0 to feel as robust as possible=2E

I'd even say that any packages which don't build for 1=2E0 should = be deleted or at least hidden=2E Maybe of channels happen we could use a st= able vs unstable channels=2E

That's my 2 cents on 1=2E0 at least=2E

On September 4, 2018 6:55:57 PM EDT, fis trivial <ybbs=2Edaans@hotmai= l=2Ecom> wrote:

Ludovic Court=C3=A8s writes:

 Hello Guix!

 I=E2=80=99ve push=
ed to guix/maintenance=2Egit a list of things that IMO we should
 do or =
might want to do for 1=2E0, with the understanding that 1=2E0 should
 ha=
ppen in 2018 (or early 2019 at the latest!)=2E  :-)

   https://git=2Esavannah=2Egnu=2Eorg/cgit/guix/maintenance=2Egit/tree=
/doc/1=2E0=2Eorg

 The list focuses on =E2=80=9Cbig item=E2=80=9D=
 features and tasks, omitting routine bug
 fixes and improvements=2E  So=
me of these items don=E2=80=99t require a lot of work
 or expertise thou=
gh, so hopefully there=E2=80=99s enough on everyone=E2=80=99s plate=2E
<=
br> Feel free to comment, volunteer, add items (but not too many!), remove<=
br> items, promote items, etc=2E  Committers should feel free to edit the f=
ile
 directly in maintenance=2Egit, especially to mark things as done=2E=
  ;-)

 Copy of the file attached below=2E

 Ludo=E2=80=99=2E
 PS: I=E2=80=99m starting the discussion but will go AFK soon after s=
ending this
     message=2E  :-)

Hi Ludovic,

=
For near future version of GUIX, maybe 1=2E1, would it be possible to:
<=
br>    Upgrade to latest GCC as base compiler

    Use gold linker as=
 default

Based on experience, newer GCC and gold linker could reduce=
 compilation time
and memory requirement dramatically=2E  It would be be=
neficial to both build
farm which have heavy load and laptop users who h=
ave very limited memory=2E

Thanks=2E
= ------6CQ62YXE3T38D9Q66QB8XOGPKAC2JZ--