Date: Thu, 27 Apr 2023 11:40:55 -0600 From: Warner Losh <imp@bsdimp.com> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: freebsd-arch <freebsd-arch@freebsd.org>, John Baldwin <jhb@freebsd.org> Subject: Re: Future of 32-bit platforms (including i386) Message-ID: <CANCZdfpDmTs71UqKnAASwBPD0PDPqUNPk8tbH6hkZ4%2B8D%2B7g8Q@mail.gmail.com> In-Reply-To: <202304271732.33RHWLrw069782@critter.freebsd.dk> References: <aaa3e005-5f72-f422-56b1-932842379e15@FreeBSD.org> <202304271732.33RHWLrw069782@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
--00000000000065fd2705fa54dc72 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Thu, Apr 27, 2023 at 11:32=E2=80=AFAM Poul-Henning Kamp <phk@phk.freebsd= .dk> wrote: > -------- > John Baldwin writes: > > > A larger question is what to do about 32-bit platforms moving forward. > > My proposal for powerpc, i386, and armv[67] is that we say publicly > > that we anticipate not supporting them in 15. > > If we do, the first two questions we will get back are: > > 1. When does 15 happen ? > Late 2025, give or take 6 months would be our recent cadence. while we might want to change this cadance, any decision here should likely assume that cadence and we can make adjustments to the plan based on a hypothetically changed cadence would bring. So we'd plan on removing the 32-bit platforms sometime in 2024 or early 2025 at the latest, but as soon as the end of this year. So you'd no longer be able to run 'main' on these platforms after a year or two. > 2. How long time will some branch of 14 be supported ? > At least until 2027 if history is a guide (de-facto is about 2 years after the next major branch) with fading levels of support. I think the 'support model' would place it around June 2028 somewhere assuming we release 14 in June. There's clearly some fuzziness here, but for planning purposes one should expect updates to trail off in late 2026 or early 2027 and critical updates stopping sometime before 2028. At least that's what I'm observing with EOL of 12 that's pending... as well as what's happened more organically for 10 and 11. Of course, the above is my opinion, and it's phrased such as to give some less vague timelines to John's proposal. Is this helpful? Warner > -- > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD committer | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by incompetenc= e. > > --00000000000065fd2705fa54dc72 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">= <div dir=3D"ltr" class=3D"gmail_attr">On Thu, Apr 27, 2023 at 11:32=E2=80= =AFAM Poul-Henning Kamp <<a href=3D"mailto:phk@phk.freebsd.dk">phk@phk.f= reebsd.dk</a>> wrote:<br></div><blockquote class=3D"gmail_quote" style= =3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding= -left:1ex">--------<br> John Baldwin writes:<br> <br> > A larger question is what to do about 32-bit platforms moving forward.= <br> > My proposal for powerpc, i386, and armv[67] is that we say publicly<br= > > that we anticipate not supporting them in 15.<br> <br> If we do, the first two questions we will get back are:<br> <br> 1. When does 15 happen ?<br></blockquote><div><br></div><div>Late 2025, giv= e or take 6 months would be our recent cadence. while we might want to chan= ge</div><div>this cadance, any decision here should likely assume that cade= nce and we can make adjustments</div><div>to the plan based on a hypothetic= ally changed cadence would bring. So we'd plan on removing</div><div>th= e 32-bit platforms sometime in 2024 or early 2025 at the latest, but as soo= n as the end of this</div><div>year. So you'd no longer be able to run = 'main' on these platforms after a year or two.<br></div><div>=C2=A0= </div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;b= order-left:1px solid rgb(204,204,204);padding-left:1ex"> 2. How long time will some branch of 14 be supported ?<br></blockquote><div= ><br></div><div>At least until 2027 if history is a guide (de-facto is abou= t 2 years after the next major branch) with</div><div>fading levels of supp= ort. I think the 'support model' would place it around June 2028 so= mewhere</div><div>assuming we release 14 in June. There's clearly some = fuzziness here, but for planning purposes</div><div>one should expect updat= es to trail off in late 2026 or early 2027 and critical updates stopping <b= r></div><div>sometime before 2028. At least that's what I'm observi= ng with EOL of 12 that's pending... as</div><div>well as what's hap= pened more organically for 10 and 11.<br></div><div><br></div><div>Of cours= e, the above is my opinion, and it's phrased such as to give some less = vague timelines to</div><div>John's proposal.</div><div><br></div><div>= Is this helpful?</div><div><br></div><div>Warner<br></div><div>=C2=A0</div>= <blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-= left:1px solid rgb(204,204,204);padding-left:1ex"> -- <br> Poul-Henning Kamp=C2=A0 =C2=A0 =C2=A0 =C2=A0| UNIX since Zilog Zeus 3.20<br= > phk@FreeBSD.ORG=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0| TCP/IP since RFC 956<br> FreeBSD committer=C2=A0 =C2=A0 =C2=A0 =C2=A0| BSD since 4.3-tahoe=C2=A0 =C2= =A0 <br> Never attribute to malice what can adequately be explained by incompetence.= <br> <br> </blockquote></div></div> --00000000000065fd2705fa54dc72--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfpDmTs71UqKnAASwBPD0PDPqUNPk8tbH6hkZ4%2B8D%2B7g8Q>