[TUHS] when did v8 or later get networking?

[TUHS] when did v8 or later get networking?

[Larry McVoy]

TCP/IP, not datakit.

[TUHS] Re: when did v8 or later get networking?

[segaloco via TUHS]

> TCP/IP, not datakit.

The V8 manual contains internet(3), tcp(3), and udp(3).  In the copy I have these are dated 8/7/85.  Down in the kernel in /usr/sys/inet the earliest files appear to be dated 3/22/85.  I'm going to do a comparison with the 4.1cBSD TCP/IP stuff because a casual glance reveals some overlap.

- Matt G.

[TUHS] Re: when did v8 or later get networking?

[Warner Losh]

[-- Attachment #1: Type: text/plain, Size: 516 bytes --]

On Wed, Aug 9, 2023, 8:38 PM segaloco via TUHS <tuhs@tuhs.org> wrote:

> > TCP/IP, not datakit.
>
> The V8 manual contains internet(3), tcp(3), and udp(3).  In the copy I
> have these are dated 8/7/85.  Down in the kernel in /usr/sys/inet the
> earliest files appear to be dated 3/22/85.  I'm going to do a comparison
> with the 4.1cBSD TCP/IP stuff because a casual glance reveals some overlap.
>

Yea, I thought it was 4.1bsd + later tcp code but with a STREAMS instead of
Socket interface...

Warner

- Matt G.
>

[-- Attachment #2: Type: text/html, Size: 1090 bytes --]

[TUHS] Re: when did v8 or later get networking?

[segaloco via TUHS]

> > > TCP/IP, not datakit.
> > 
> > The V8 manual contains internet(3), tcp(3), and udp(3). In the copy I have these are dated 8/7/85. Down in the kernel in /usr/sys/inet the earliest files appear to be dated 3/22/85. I'm going to do a comparison with the 4.1cBSD TCP/IP stuff because a casual glance reveals some overlap.
> > - Matt G.
> 
> 
> Yea, I thought it was 4.1bsd + later tcp code but with a STREAMS instead of Socket interface...
> 
> Warner

This comment at the top of /usr/sys/netinet/socket.c is illuminating:

> /*
>  * socket emulation routines
>  * pretty much 4.2 sys/uipc_socket2.c
>  */

All of the files that have timestamps at the top list 83/07/29, except ip_input.c which has 83/08/16 instead.  The V8 version has _device (device driver) and _ld (line discipline) components that the 4.1cBSD code does not have.  Many other files have analogs between the two.  The byte ordering subroutines have been copied into a file, goo.s, from their home in 4.1cBSD in the C library (/usr/src/lib/libc/net/misc).  When this work originated someone else would need to answer, but looks like it was as Warner described, modified 4.xBSD TCP/IP.

- Matt G.

[TUHS] Re: when did v8 or later get networking?

[Rob Pike]

[-- Attachment #1: Type: text/plain, Size: 842 bytes --]

STREAMS was the Unix support group's shouty adaptation. The V8 one that dmr
put together was the gentler dulcet-toned Streams.

Seriously, no one could figure out why they wanted to yell.

-rob


On Thu, Aug 10, 2023 at 12:46 PM Warner Losh <imp@bsdimp.com> wrote:

>
>
> On Wed, Aug 9, 2023, 8:38 PM segaloco via TUHS <tuhs@tuhs.org> wrote:
>
>> > TCP/IP, not datakit.
>>
>> The V8 manual contains internet(3), tcp(3), and udp(3).  In the copy I
>> have these are dated 8/7/85.  Down in the kernel in /usr/sys/inet the
>> earliest files appear to be dated 3/22/85.  I'm going to do a comparison
>> with the 4.1cBSD TCP/IP stuff because a casual glance reveals some overlap.
>>
>
> Yea, I thought it was 4.1bsd + later tcp code but with a STREAMS instead
> of Socket interface...
>
> Warner
>
> - Matt G.
>>
>

[-- Attachment #2: Type: text/html, Size: 2117 bytes --]

[TUHS] Re: when did v8 or later get networking?

[John Cowan]

[-- Attachment #1: Type: text/plain, Size: 365 bytes --]

On Wed, Aug 9, 2023 at 11:18 PM Rob Pike <robpike@gmail.com> wrote:

> STREAMS was the Unix support group's shouty adaptation. The V8 one that
> dmr put together was the gentler dulcet-toned Streams.
>
> Seriously, no one could figure out why they wanted to yell.
>

I always supposed it was a trademark (like UNIX), or at least was intended
to be one.

[-- Attachment #2: Type: text/html, Size: 1039 bytes --]

[TUHS] Re: when did v8 or later get networking?

[Douglas McIlroy]

My preface to the manual: says v8 " incorporates facilities from the
Seventh Edition, System V, andBerkeley BSD 4.1. The distinctive theme
...is distributed computing"It goes on to cite streams, Datakit, RFS,
and Blit. The internet lurks in the citation of BSD 4.1.

The preface goes on to suggest how the new facilities affected life in
the lab: "These facilities have spurred a host of programs for remote
information services, interactive graphics, debugging, and simple
fun".

On rereading this, I am struck by how blinkered I was not to overtly
mention either the internet, which brought us out of the clumsy era of
uucp and CSNET, or upas, which seamlessly integrated remote and local
mail (and would eventually save us from the Morris worm).

I recall expressions of surprise (dismay?) at the size of the BSD
internet code, but without it we'd have lost our place in the Unix
community. Even greater concern about the complexity of sendmail led
Presotto to develop the simpler upas.

Doug


On Wed, Aug 9, 2023 at 9:09 PM Larry McVoy <lm@mcvoy.com> wrote:
>
>
> TCP/IP, not datakit.

[TUHS] Re: when did v8 or later get networking?

[Jonathan Gray]

On Wed, Aug 09, 2023 at 06:09:40PM -0700, Larry McVoy wrote:
> 
> TCP/IP, not datakit.
> 

"The closest we got to using 4.n BSD was when Robert Morris, now at MIT,
imported the 4.1c TCP/IP stack into 7/8th edition (I believe in 84)
nominally as my summer student."
Dave Presotto on 9fans, 2001-10-01
https://marc.info/?l=9fans&m=111558818409805&w=2

"the 8th edition system including Streams was already pretty
much in place by the time that Robert Morris adapted
the then-current BSD TCP/IP stack to streams.  At that time,
we were using either serial communication over various modems,
and more notably Datakit. Looking back at this, one of
the satisfying things is that the communication structure
built then was adaptable so smoothly to TCP/IP when
the protocol's importance became undeniable."
Dennis Ritchie on 9fans, 2001-10-02
https://marc.info/?l=9fans&m=111558818709848&w=2