Date: Fri, 28 Apr 2023 21:42:09 +0300 From: Mehmet Erol Sanliturk <m.e.sanliturk@gmail.com> To: Hans Petter Selasky <hps@selasky.org> Cc: Jessica Clarke <jrtc27@freebsd.org>, freebsd-arch <freebsd-arch@freebsd.org>, John Baldwin <jhb@freebsd.org> Subject: Re: Future of 32-bit platforms (including i386) Message-ID: <CAOgwaMsicnR0=sEVAXFHdRwA3WDB6gi_=hG7Q6N2kAD7KVUe3Q@mail.gmail.com> In-Reply-To: <6079003b-7df2-f3c2-f624-6fe39a1cf9c0@selasky.org> References: <aaa3e005-5f72-f422-56b1-932842379e15@FreeBSD.org> <671d3bf6-b207-e7c5-5282-4df317193db6@selasky.org> <A260039D-B51F-4593-9A85-8A5CF2818DEF@freebsd.org> <6079003b-7df2-f3c2-f624-6fe39a1cf9c0@selasky.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--000000000000cb135b05fa69d6d5 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, Apr 28, 2023 at 8:45=E2=80=AFPM Hans Petter Selasky <hps@selasky.or= g> wrote: > On 4/28/23 01:50, Jessica Clarke wrote: > > On 28 Apr 2023, at 00:44, Hans Petter Selasky <hps@selasky.org> wrote: > >> > >> On 4/27/23 19:19, John Baldwin wrote: > >>> For 13.0, i386 was demoted from Tier 1 to Tier 2. In the announcemen= t > >>> of this for 13.0, the project committed to an update on i386's future > >>> around the time of 14.0. The announcement at the time suggested that > >>> i386 would be supported less in 14.x than in 13.x. > >> > >> Hi, > >> > >> This makes me think about all the issues about the "long" type in the > past, and printf() and more, being caught when compiling TARGET_ARCH=3Di3= 86 . > >> > >> Maybe just put the following line of code somewhere central :-) > >> > >> _Static_assert(sizeof(long) =3D=3D 8); > >> > >> Will there ever be some kind of hybrid CPU systems? > >> > >> 4 cores AMD64, 4 cores AARCH64 and some virtual QEMU CPUs all running > on the same system? > >> > >> I mean, the arm vs intel battle is not going to end soonish. And > emulating CPUs is slow and waste electricity. Why not have one computer > having both kind of CPUs, and one OS, and one harddisk? And figure out a > common ABI allowing seamless task switching between them? I know there ar= e > some hard differences, but can't those be ironed out? > > > > I don=E2=80=99t know where to start with this other than to give an emp= hatic no > to almost all of what you said, or at least the bits for which meaning ca= n > be extracted. Regardless, this is not the place for such pie-in-the-sky > discussions; if you want to theorise about weird and wacky computer > architectures then please take it elsewhere. > > > > Hi Jess, > > I'd like to know why you think this is a wacky idea, to have a super-set > computer architecture, where each CPU can run the full instruction set > of both ARM64 and AARCH64 at the same time. > > You have an open invitation for a video call on FaceBook or whatever you > prefer to talk about this. Send me something off-list. > > --HPS > > It is not necessary to go to a very far distant future . Assume you have a cluster of boards with different CPUs . Then schedule execution of your programs with respect to the required CPU on this cluster . Is this possible with FreeBSD ? Is it a good or bad idea to have such a facility ? Mehmet Erol Sanliturk --000000000000cb135b05fa69d6d5 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div dir=3D"ltr"><div class=3D"gmail_default" style=3D"fon= t-family:monospace;font-size:large"><br></div></div><br><div class=3D"gmail= _quote"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Apr 28, 2023 at 8:45= =E2=80=AFPM Hans Petter Selasky <<a href=3D"mailto:hps@selasky.org">hps@= selasky.org</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">On 4/28/23 01:50, Jessica Clarke wrote:<br> > On 28 Apr 2023, at 00:44, Hans Petter Selasky <<a href=3D"mailto:hp= s@selasky.org" target=3D"_blank">hps@selasky.org</a>> wrote:<br> >><br> >> On 4/27/23 19:19, John Baldwin wrote:<br> >>> For 13.0, i386 was demoted from Tier 1 to Tier 2.=C2=A0 In the= announcement<br> >>> of this for 13.0, the project committed to an update on i386&#= 39;s future<br> >>> around the time of 14.0.=C2=A0 The announcement at the time su= ggested that<br> >>> i386 would be supported less in 14.x than in 13.x.<br> >><br> >> Hi,<br> >><br> >> This makes me think about all the issues about the "long"= ; type in the past, and printf() and more, being caught when compiling TARG= ET_ARCH=3Di386 .<br> >><br> >> Maybe just put the following line of code somewhere central :-)<br= > >><br> >> _Static_assert(sizeof(long) =3D=3D 8);<br> >><br> >> Will there ever be some kind of hybrid CPU systems?<br> >><br> >> 4 cores AMD64, 4 cores AARCH64 and some virtual QEMU CPUs all runn= ing on the same system?<br> >><br> >> I mean, the arm vs intel battle is not going to end soonish. And e= mulating CPUs is slow and waste electricity. Why not have one computer havi= ng both kind of CPUs, and one OS, and one harddisk? And figure out a common= ABI allowing seamless task switching between them? I know there are some h= ard differences, but can't those be ironed out?<br> > <br> > I don=E2=80=99t know where to start with this other than to give an em= phatic no to almost all of what you said, or at least the bits for which me= aning can be extracted. Regardless, this is not the place for such pie-in-t= he-sky discussions; if you want to theorise about weird and wacky computer = architectures then please take it elsewhere.<br> > <br> <br> Hi Jess,<br> <br> I'd like to know why you think this is a wacky idea, to have a super-se= t <br> computer architecture, where each CPU can run the full instruction set <br> of both ARM64 and AARCH64 at the same time.<br> <br> You have an open invitation for a video call on FaceBook or whatever you <b= r> prefer to talk about this. Send me something off-list.<br> <br> --HPS<br> <br></blockquote><div><br></div><div><div style=3D"font-family:monospace;fo= nt-size:large" class=3D"gmail_default">It is not necessary to go to a very = far distant future .<br></div><div style=3D"font-family:monospace;font-size= :large" class=3D"gmail_default"><br></div><div style=3D"font-family:monospa= ce;font-size:large" class=3D"gmail_default">Assume you have a cluster of bo= ards with different CPUs .</div><div style=3D"font-family:monospace;font-si= ze:large" class=3D"gmail_default">Then schedule execution of your programs = with respect to the required CPU on this cluster .</div><div style=3D"font-= family:monospace;font-size:large" class=3D"gmail_default"><br></div><div st= yle=3D"font-family:monospace;font-size:large" class=3D"gmail_default">Is th= is possible with FreeBSD ?</div><div style=3D"font-family:monospace;font-si= ze:large" class=3D"gmail_default">Is it a good or bad idea to have such a f= acility ?</div><div style=3D"font-family:monospace;font-size:large" class= =3D"gmail_default"><br></div><div style=3D"font-family:monospace;font-size:= large" class=3D"gmail_default"><br></div><div style=3D"font-family:monospac= e;font-size:large" class=3D"gmail_default"><br></div><div style=3D"font-fam= ily:monospace;font-size:large" class=3D"gmail_default"><br></div><div style= =3D"font-family:monospace;font-size:large" class=3D"gmail_default">Mehmet E= rol Sanliturk<br></div><div style=3D"font-family:monospace;font-size:large"= class=3D"gmail_default"></div><br></div><div><br></div><div>=C2=A0</div></= div></div> --000000000000cb135b05fa69d6d5--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAOgwaMsicnR0=sEVAXFHdRwA3WDB6gi_=hG7Q6N2kAD7KVUe3Q>