Re: [9fans] Nemo's Opus
To add to the previous comments, I'm a huge fan of Nemo's “Introduction to Operating Systems Abstractions”: https://lsub.org/who/nemo/9.intro.pdf It's just a brilliant guide to finding your way around the system, and doing some programming in it, especially if you're not a kernel-head and talk of "ring 0" makes your brain hurt. Not that my being a huge fan of anything of anything is in general a recommendation, but in this case I'm right. On 10 September 2016 at 07:36, Sergey Zhilkinwrote: > 5 cents from me, as I'm fan of Nemo's writings (and code also) link to his > papers https://lsub.org/who/nemo/papers.html > > 2016-09-09 18:54 GMT+03:00 Brantley Coile : > >> I’ve been reading Nemo’s “Notes on the Plan 9 3rd edition Kernel Source” >> after a number of years. Three things struck me on this reading of what is >> a great and much appreciated work. First, is what a good job Francisco did >> with this work. Even though he never finished it, having been overtaking, I >> think, by the 4th edition, it is a very good introduction to an operating >> system suitable for instruction in a undergraduate or graduate class in >> operating systems. I, fortunately, don’t have to teach, but if I did, I >> would certainly use the work. >> >> Second, I’m struck by how much larger the system had grown by the time >> Nemo wrote the commentary. I had the good fortune to read John Lion’s >> commentary on Plan 9 during my brief tenure at Bell Labs in 1990. If I >> remember right, the kernel I was using was bout 25,000 lines. The first >> version I used outside the Labs was the 2nd edition it weighs in at a hefty >> 39,000 lines. The current system I’m running, the 32 bit one, not the 64 >> bit one, is 140,000 lines. I’m not sure the size of the 3rd edition, but >> the growth is interesting. >> >> The third thing that struct me is the changes in the Intel architecture >> since the original PC based port. The first Plan 9 for PC ran on AT 386 >> machines in the 1990’s, if I remember right. Those were the days of ISA and >> EISA and before PCI made it’s plug-and-play appearance on the scene. It >> seems that while the PC stuff has kept up with most of the many changes in >> the Intel hardware platform over the years, there is still some cruft from >> the old days. >> >> All very interesting to think about. I highly recommend Nemo’s book. >> Here’s a link to it. >> >> http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.75. >> 5409=rep1=pdf >> >> Brantley Coile >> b...@coraid.com >> http://coraid.com >> >> >> > > > -- > С наилучшими пожеланиями > Жилкин Сергей > With best regards > Zhilkin Sergey >
Re: [9fans] Distros
>> he wrote *AMD*64 not *ARM*64 ;-) > > Does plan9 on the rpi run armv8? I thought it was armv7 even on the rpi3. it was a joke... -- cinap
Re: [9fans] Distros
2016-09-11 19:50 GMT+02:00: > he wrote *AMD*64 not *ARM*64 ;-) Does plan9 on the rpi run armv8? I thought it was armv7 even on the rpi3. Andrés
Re: [9fans] Distros
2016-09-11 19:48 GMT+02:00 Kurt H Maier: > On Sun, Sep 11, 2016 at 07:39:31PM +0200, Andrés Domínguez wrote: >> 2016-09-11 18:40 GMT+02:00 Benjamin Huntsman >> : >> > with amd64 support? >> >> the rpi image > > I predict suffering. I understood that he wanted to start with a distribution with amd64 support, not to start running it on amd64. Most, if not all, rpi sources has been integrated into bell labs' distribution. It's not clear to me if he wants the distribution with better amd64 support or the most used/developed/supported with amd64 support (from the question looks more the later). Andrés
Re: [9fans] Distros
he wrote *AMD*64 not *ARM*64 ;-) -- cinap
Re: [9fans] Distros
On Sun, Sep 11, 2016 at 07:39:31PM +0200, Andrés Domínguez wrote: > 2016-09-11 18:40 GMT+02:00 Benjamin Huntsman >: > > with amd64 support? > > the rpi image > I predict suffering. Benjamin: I wouldn't get too hung up on amd64 specifically. Labs, 9atom, and 9front should all work fine out of the box, and by the time you're using the system enough to be interested in accessing larger memory, you'll have a clear idea of which direction you want to go. I'm making an assumption there that memory support is what motivates your request for amd64 -- is there another reason you have in mind? For most cases, there isn't a tremendous difference a user would notice otherwise. i386 and amd64 live in harmony here. khm
Re: [9fans] Distros
2016-09-11 18:40 GMT+02:00 Benjamin Huntsman: > Hi! > >What would be the best place to start in, looking for the most > current/widely-used distribution with amd64 support? 9front? 9atom? > Other? I would bet that the best distribution to start working with out of the box is the rpi image by Richard Miller. It's bell labs' based, afaik, but probably the most active fork is 9front (or maybe there is no fork). Andrés
[9fans] 9front on raspberry pi
Hi All, I have an experimental (read unstable, may void the warranty) merge of Richard Miller's kernel for raspberry Pi into 9front. It is capable of running Go programs and sdk. The code is here as a drop in replacement for existing /sys/src/9/bcm: https://github.com/sirnewton01/rpi-9front Also, there is a release with an sdcard image there from the last 9front release. Tested briefly with a Pi B+ and Pi 2 B. In theory it could work with other models such as zero. If you find problems either with the code or in testing comments are welcome either as response to this email or github issue. Pull requests are welcome. Cheers, Chris
[9fans] Distros
Hi! Since a similar line of conversation came up recently, I thought I'd ask... Maybe this was just my perception, but I remember that years ago, there seemed to be a pretty strong aversion to forking the official distribution from Bell Labs. These days I'm not certain where the majority of the activity occurs, and which of the forks is the most "official", or at least "mainstream"... What would be the best place to start in, looking for the most current/widely-used distribution with amd64 support? 9front? 9atom? Other? As always, many thanks! -Ben