Re: Reviewing Hurd-on-L4 (and considering its future?)
Rick Hodgin <[email protected]> Wed, 23 May 2018 18:06:21 +0000 (UTC)
| Newsgroups | gmane.os.hurd.l4 |
|---|---|
| Message-ID | <[email protected]> |
------=_Part_4632815_1009118744.1527098781101 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Forwarding my two replies sent directly to Paul and RMS to the list. --=C2=A0Rick C. Hodgin ----- Forwarded Message ----- From: Rick Hodgin <[email protected]>= To: Paul Boddie <[email protected]>Cc: Richard Stallman <[email protected]>Sent:= Wednesday, May 23, 2018, 2:03:23 PM EDTSubject: Re: Reviewing Hurd-on-L4 (= and considering its future?) Paul, Here is the part of the original email chain (we had several back-and-forth= ).I had written him asking for his advice on replacing the HURD kernel.=C2= =A0 Ican't find my original sent emails ... I may have purged or archived t= hemat some point, only these replies: -----[ Begin original email ]------------- Original Message -------- From: Richard Stallman <[email protected]> Sent: Tue, Jun 19, 2012 06:21 PM To: Rick C. Hodgin <[email protected]> CC:=C2=A0 Subject: Re: Replacement for the HURD =C2=A0 =C2=A0 With either Linux or the HURD (as the HURD stands today) we c= annot ever=C2=A0 =C2=A0 =C2=A0 get there.=C2=A0 We cannot ever get to where we have a truly = free operating=C2=A0 =C2=A0 =C2=A0 system (kernel and up), even if we have the outward appearanc= e and=C2=A0 =C2=A0 =C2=A0 illusion of it. I don't think that follows.=C2=A0 Both Linux and Hurd _are_ free kernels. The problem is that we lack free drivers and firmware for many peripherals. Replacing the free code of Linux with some other free code won't get us the drivers or the firmware.=C2=A0 That requires writing drivers and fir= mware. I simply don't see how replacing the free code of Linux with other free cod= e would make progress along that dimension. =C2=A0 =C2=A0 I realized at that time how easily I could write a competitor= to=C2=A0 =C2=A0 =C2=A0 Blender, or a 3D CAD system.=C2=A0 I even looked into it a li= ttle bit, but=C2=A0 =C2=A0 =C2=A0 saw how many CAD programs there were out there and thought I = could not=C2=A0 =C2=A0 =C2=A0 write one sufficiently better than what's available to make i= t worth the=C2=A0 =C2=A0 =C2=A0 amount of time it would require. The existing free CAD programs have a license incompatibility with GPLv3, so it would be very useful to write a replacement.=C2=A0 However, one of th= em is trying to eliminate its license incompatibility, and that helping that project do so would be very useful. --=C2=A0 Dr Richard Stallman President, Free Software Foundation 51 Franklin St Boston MA 02110 USA www.fsf.org=C2=A0 www.gnu.org Skype: No way! That's nonfree (freedom-denying) software. =C2=A0 Use Ekiga or an ordinary phone call -----[ End original email ]----- --=C2=A0Rick C. Hodgin On Wednesday, May 23, 2018, 1:54:45 PM EDT, Rick Hodgin <foxmuldrster@y= ahoo.com> wrote: Paul, I contacted Richard Stallman in 2012 and asked him about continuingdevelopm= ent on the HURD kernel.=C2=A0 He told me that it wasn't really need-ed anym= ore because we have Linux.=C2=A0 I tried to convince him that GNUneeded its= own kernel and we shouldn't rely upon Linux, but he toldme that instead of= spending time working on a GNU kernel, I shouldmove to work on making a co= mpatible Adobe reader, and other suchpractical applications. I don't think there's any interest from Stallman down on working on aGNU ke= rnel.=C2=A0 I think that's a mistake, by the way, but to my knowledgethat's= where we're at. I've cc'd Stallman on this email so he can verify his position. --=C2=A0Rick C. Hodgin On Wednesday, May 23, 2018, 1:02:55 PM EDT, Paul Boddie <[email protected]= g.uk> wrote: Hello, The day before yesterday, I found myself editing the L4 page on the Debian= =20 Wiki: https://wiki.debian.org/L4 It occurred to me that in the context of the Hurd, very few updates about L= 4=20 implementations have been propagated via various channels for probably a=20 decade or more. This meant that the above page was outdated as well as=20 inaccurate or incorrect on some matters (and it wasn't exactly a long page,= =20 either). Reviewing the archives for the l4-hurd list, I see that very little discuss= ion=20 has taken place since 2010 or so, where Viengoos was once again mentioned. = And=20 the background to all this is, of course, provided on the Hurd site: https://www.gnu.org/software/hurd/history/port_to_another_microkernel.html What I would like to know is whether more modern L4 implementations might b= e=20 reasonable candidates for attempting this kind of work again. It seems to m= e=20 that L4 implementations of the early 2000s were considered inappropriate or= =20 insufficient for the design of "Hurd-NG", leading to things like Viengoos. But then again, L4 implementations have since been widely deployed, albeit= =20 perhaps in very specific roles. Meanwhile, environments like L4Re and Genod= e=20 must surely go some of the way towards providing the basis of something lik= e=20 the Hurd. The hurd-l4 work that was previously released got as far as running the ban= ner=20 program, apparently, but surely things like L4Re and Genode go quite a bit= =20 further than that. Obviously, they are merely the foundations of any potent= ial=20 Hurd-like system, but it would not be unfair to claim that anyone wanting t= o=20 implement such a system would start off with a lot more code already writte= n=20 than was the case fifteen or so years ago. So, has anyone seriously considered a fresh approach to this endeavour usin= g=20 more recent L4 implementations and environments? I largely ask because I ha= ve=20 been developing programs to run on certain niche hardware devices using L4R= e=20 and the Fiasco.OC microkernel, and although some aspects of this are=20 frustrating, it is surprising how much supporting code there is already. I = can=20 envisage building components on top that support a more general computing= =20 environment. Does anyone have any constructive perspectives on this? Paul =20 ------=_Part_4632815_1009118744.1527098781101 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <html><head></head><body><div style=3D"font-family:Helvetica Neue, Helvetic= a, Arial, sans-serif;font-size:13px;"><div></div> <div>Forwarding my two replies sent directly to Paul and RMS to= the list.</div><div><br></div><div>-- </div><div>Rick C. Hodgin</div>= <div><br></div> =20 <div id=3D"ydp11aeebfbyahoo_quoted_7900444259" class=3D"ydp11ae= ebfbyahoo_quoted"> <div style=3D"font-family:'Helvetica Neue', Helvetica, Aria= l, sans-serif;font-size:13px;color:#26282a;"> <div>----- Forwarded Message -----</div> <div><b>From:</b> Rick Hodgin <[email protected]= m></div><div><b>To:</b> Paul Boddie <[email protected]></div><div= ><b>Cc:</b> Richard Stallman <[email protected]></div><div><b>Sent:</b> Wed= nesday, May 23, 2018, 2:03:23 PM EDT</div><div><b>Subject:</b> Re: Reviewin= g Hurd-on-L4 (and considering its future?)</div><div><br></div> <div><div id=3D"ydp11aeebfbyiv9154147894"><div><div sty= le=3D"font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:13= px;"><div></div> <div>Paul,</div><div><br clear=3D"none"></div><div>Here is the = part of the original email chain (we had several back-and-forth).</div><div= >I had written him asking for his advice on replacing the HURD kernel. = ; I</div><div>can't find my original sent emails ... I may have purged or a= rchived them</div><div>at some point, only these replies:</div><div><br cle= ar=3D"none"></div><div>-----[ Begin original email ]-----</div><div><span><= span>-------- Original Message --------</span><br clear=3D"none"><span>From= : Richard Stallman <</span><a shape=3D"rect" href=3D"mailto:[email protected]"= style=3D"color:rgb(25, 106, 212);text-decoration-line:underline;" rel=3D"n= ofollow" target=3D"_blank">[email protected]</a><span>></span><br clear=3D"non= e"><span>Sent: Tue, Jun 19, 2012 06:21 PM</span><br clear=3D"none"><span>To= : Rick C. Hodgin <</span><a shape=3D"rect" href=3D"mailto:foxmuldrster@y= ahoo.com" style=3D"color:rgb(25, 106, 212);text-decoration-line:underline;"= rel=3D"nofollow" target=3D"_blank">[email protected]</a><span>></s= pan><br clear=3D"none"><span>CC: </span><br clear=3D"none"><span>Subje= ct: Re: Replacement for the HURD</span><br clear=3D"none"><br clear=3D"none= "><span> With either Linux or the HURD (as the HURD stands tod= ay) we cannot ever </span><br clear=3D"none"><span> get t= here. We cannot ever get to where we have a truly free operating = ;</span><br clear=3D"none"><span> system (kernel and up), even= if we have the outward appearance and </span><br clear=3D"none"><span= > illusion of it.</span><br clear=3D"none"><br clear=3D"none">= <span>I don't think that follows. Both Linux and Hurd _are_ free kern= els.</span><br clear=3D"none"><span>The problem is that we lack free driver= s and firmware for many peripherals.</span><br clear=3D"none"><br clear=3D"= none"><span>Replacing the free code of Linux with some other free code won'= t get</span><br clear=3D"none"><span>us the drivers or the firmware. = That requires writing drivers and firmware.</span><br clear=3D"none"><br cl= ear=3D"none"><span>I simply don't see how replacing the free code of Linux = with other free code</span><br clear=3D"none"><span>would make progress alo= ng that dimension.</span><br clear=3D"none"><br clear=3D"none"><span> = I realized at that time how easily I could write a competitor to&nb= sp;</span><br clear=3D"none"><span> Blender, or a 3D CAD syste= m. I even looked into it a little bit, but </span><br clear=3D"n= one"><span> saw how many CAD programs there were out there and= thought I could not </span><br clear=3D"none"><span> wri= te one sufficiently better than what's available to make it worth the = </span><br clear=3D"none"><span> amount of time it would requi= re.</span><br clear=3D"none"><br clear=3D"none"><span>The existing free CAD= programs have a license incompatibility with GPLv3,</span><br clear=3D"non= e"><span>so it would be very useful to write a replacement. However, = one of them</span><br clear=3D"none"><span>is trying to eliminate its licen= se incompatibility, and that helping</span><br clear=3D"none"><span>that pr= oject do so would be very useful.</span><br clear=3D"none"><br clear=3D"non= e"><span>-- </span><br clear=3D"none"><span>Dr Richard Stallman</span>= <br clear=3D"none"><span>President, Free Software Foundation</span><br clea= r=3D"none"><span>51 Franklin St</span><br clear=3D"none"><span>Boston MA 02= 110</span><br clear=3D"none"><span>USA</span><br clear=3D"none"><span>www.f= sf.org www.gnu.org</span><br clear=3D"none"><span>Skype: No way! That= 's nonfree (freedom-denying) software.</span><br clear=3D"none"><span> = ; Use Ekiga or an ordinary phone call</span></span><br clear=3D"none"></div= ><div>-----[ End original email ]-----</div><div><br clear=3D"none"></div><= div>-- </div><div>Rick C. Hodgin</div><div><br clear=3D"none"></div><d= iv><br clear=3D"none"></div> =20 <div class=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yahoo_quoted" = id=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yahoo_quoted_8045459901"> <div style=3D"font-family:'Helvetica Neue', Helvetica, Aria= l, sans-serif;font-size:13px;color:#26282a;"> =20 <div class=3D"ydp11aeebfbyiv9154147894yqt4210211962" id= =3D"ydp11aeebfbyiv9154147894yqt05314"><div> On Wednesday, May 23, 2018, 1:54:45 PM EDT, Rick Ho= dgin <[email protected]> wrote: </div> <div>Paul,<br clear=3D"none"></div></div><div><div id= =3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yiv5249886636"><div style=3D"font-fa= mily:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:13px;"><div cla= ss=3D"ydp11aeebfbyiv9154147894yqt4210211962" id=3D"ydp11aeebfbyiv9154147894= yqt10288"><div><br clear=3D"none"></div><div>I contacted Richard Stallman i= n 2012 and asked him about continuing</div><div>development on the HURD ker= nel. He told me that it wasn't really need-</div><div>ed anymore beca= use we have Linux. I tried to convince him that GNU</div><div>needed = its own kernel and we shouldn't rely upon Linux, but he told</div><div>me t= hat instead of spending time working on a GNU kernel, I should</div><div>mo= ve to work on making a compatible Adobe reader, and other such</div><div>pr= actical applications.</div><div><br clear=3D"none"></div><div>I don't think= there's any interest from Stallman down on working on a</div><div>GNU kern= el. I think that's a mistake, by the way, but to my knowledge</div><d= iv>that's where we're at.</div><div><br clear=3D"none"></div><div>I've cc'd= Stallman on this email so he can verify his position.</div><div><br clear= =3D"none"></div><div>-- </div><div>Rick C. Hodgin</div><div><br clear= =3D"none"></div></div> =20 <div class=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yiv5249886636y= dpcfec05c7yahoo_quoted" id=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yiv5249886= 636ydpcfec05c7yahoo_quoted_7464239462"> <div style=3D"font-family:'Helvetica Neue', Helvetica, Aria= l, sans-serif;font-size:13px;color:#26282a;"> =20 <div class=3D"ydp11aeebfbyiv9154147894yqt4210211962" id= =3D"ydp11aeebfbyiv9154147894yqt20832"><div class=3D"ydp11aeebfbyiv915414789= 4ydpdfa62cd5yiv5249886636yqt1468248884" id=3D"ydp11aeebfbyiv9154147894ydpdf= a62cd5yiv5249886636yqt17087"><div> On Wednesday, May 23, 2018, 1:02:55 PM EDT, Paul Bo= ddie <[email protected]> wrote: </div> <div>Hello,<br clear=3D"none"></div></div></div><div><d= iv class=3D"ydp11aeebfbyiv9154147894yqt4210211962" id=3D"ydp11aeebfbyiv9154= 147894yqt37733"><div class=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yiv5249886= 636yqt1468248884" id=3D"ydp11aeebfbyiv9154147894ydpdfa62cd5yiv5249886636yqt= 14552"><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr">The day b= efore yesterday, I found myself editing the L4 page on the Debian <br clear= =3D"none"></div><div dir=3D"ltr">Wiki:<br clear=3D"none"></div><div dir=3D"= ltr"><br clear=3D"none"></div><div dir=3D"ltr"><a shape=3D"rect" href=3D"ht= tps://wiki.debian.org/L4" rel=3D"nofollow" target=3D"_blank">https://wiki.d= ebian.org/L4</a><br clear=3D"none"></div><div dir=3D"ltr"><br clear=3D"none= "></div><div dir=3D"ltr">It occurred to me that in the context of the Hurd,= very few updates about L4 <br clear=3D"none"></div><div dir=3D"ltr">implem= entations have been propagated via various channels for probably a <br clea= r=3D"none"></div><div dir=3D"ltr">decade or more. This meant that the above= page was outdated as well as <br clear=3D"none"></div><div dir=3D"ltr">ina= ccurate or incorrect on some matters (and it wasn't exactly a long page, <b= r clear=3D"none"></div><div dir=3D"ltr">either).<br clear=3D"none"></div><d= iv dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr">Reviewing the arch= ives for the l4-hurd list, I see that very little discussion <br clear=3D"n= one"></div><div dir=3D"ltr">has taken place since 2010 or so, where Viengoo= s was once again mentioned. And <br clear=3D"none"></div><div dir=3D"ltr">t= he background to all this is, of course, provided on the Hurd site:<br clea= r=3D"none"></div><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr"= ><a shape=3D"rect" href=3D"https://www.gnu.org/software/hurd/history/port_t= o_another_microkernel.html" rel=3D"nofollow" target=3D"_blank">https://www.= gnu.org/software/hurd/history/port_to_another_microkernel.html</a><br clear= =3D"none"></div><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr">= What I would like to know is whether more modern L4 implementations might b= e <br clear=3D"none"></div><div dir=3D"ltr">reasonable candidates for attem= pting this kind of work again. It seems to me <br clear=3D"none"></div><div= dir=3D"ltr">that L4 implementations of the early 2000s were considered ina= ppropriate or <br clear=3D"none"></div><div dir=3D"ltr">insufficient for th= e design of "Hurd-NG", leading to things like Viengoos.<br clear=3D"none"><= /div><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr">But then ag= ain, L4 implementations have since been widely deployed, albeit <br clear= =3D"none"></div><div dir=3D"ltr">perhaps in very specific roles. Meanwhile,= environments like L4Re and Genode <br clear=3D"none"></div><div dir=3D"ltr= ">must surely go some of the way towards providing the basis of something l= ike <br clear=3D"none"></div><div dir=3D"ltr">the Hurd.<br clear=3D"none"><= /div><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"ltr">The hurd-l4= work that was previously released got as far as running the banner <br cle= ar=3D"none"></div><div dir=3D"ltr">program, apparently, but surely things l= ike L4Re and Genode go quite a bit <br clear=3D"none"></div><div dir=3D"ltr= ">further than that. Obviously, they are merely the foundations of any pote= ntial <br clear=3D"none"></div><div dir=3D"ltr">Hurd-like system, but it wo= uld not be unfair to claim that anyone wanting to <br clear=3D"none"></div>= <div dir=3D"ltr">implement such a system would start off with a lot more co= de already written <br clear=3D"none"></div><div dir=3D"ltr">than was the c= ase fifteen or so years ago.<br clear=3D"none"></div><div dir=3D"ltr"><br c= lear=3D"none"></div><div dir=3D"ltr">So, has anyone seriously considered a = fresh approach to this endeavour using <br clear=3D"none"></div><div dir=3D= "ltr">more recent L4 implementations and environments? I largely ask becaus= e I have <br clear=3D"none"></div><div dir=3D"ltr">been developing programs= to run on certain niche hardware devices using L4Re <br clear=3D"none"></d= iv><div dir=3D"ltr">and the Fiasco.OC microkernel, and although some aspect= s of this are <br clear=3D"none"></div><div dir=3D"ltr">frustrating, it is = surprising how much supporting code there is already. I can <br clear=3D"no= ne"></div><div dir=3D"ltr">envisage building components on top that support= a more general computing <br clear=3D"none"></div><div dir=3D"ltr">environ= ment.<br clear=3D"none"></div><div dir=3D"ltr"><br clear=3D"none"></div><di= v dir=3D"ltr">Does anyone have any constructive perspectives on this?<br cl= ear=3D"none"></div><div dir=3D"ltr"><br clear=3D"none"></div><div dir=3D"lt= r">Paul<br clear=3D"none"></div></div></div><div dir=3D"ltr"><br clear=3D"n= one"></div></div> </div> </div></div></div></div> </div> </div></div></div></div></div> </div> </div></div></body></html> ------=_Part_4632815_1009118744.1527098781101--