From: "Theodore Y. Ts'o" <tytso@mit.edu>
To: Clem Cole <clemc@ccc.com>
Cc: The Eunuchs Hysterical Society <tuhs@tuhs.org>,
Bakul Shah <bakul@iitbombay.org>
Subject: Re: [TUHS] 386BSD released
Date: Fri, 16 Jul 2021 11:44:26 -0400 [thread overview]
Message-ID: <YPGpWh60wFePL1en@mit.edu> (raw)
In-Reply-To: <CAC20D2O=LuFMFdShXidPyJDTZvfbdJb9DN5ouXTBA-ScFtwNJA@mail.gmail.com>
On Fri, Jul 16, 2021 at 10:40:56AM -0400, Clem Cole wrote:
>
> A huge difference, as Ted I'm sure knows, is that you tended to have many
> more serial lines than network interfaces. I suspect Rob's scheme
> would have sucked trying to support traditional single-byte serial
> interfaces or really just use too much memory to be practical.
Network interfaces tend to be much faster than serial lines; at least
an order of magnitude. And with network interfaces you care about the
packet boundaries, and you want to process each packet separately. So
that makes things a lot harder than with serial interfaces.
With serial ports, 8k per serial port is plenty (2 x 2k flip buffers,
plus a 4k tty buffer between the mid-layer and userspace) for the
receive path. On the PDP-11, memory was much more constrained, so the
clist with each cblock storing 6 characters at a time in a linked list
was probably necessary. But even in the early days of the 386, you
could afford to make a different memory/performance tradeoff.
- Ted
next prev parent reply other threads:[~2021-07-16 15:44 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-13 22:28 Dave Horsfall
2021-07-14 7:54 ` Michael Kjörling
2021-07-14 8:19 ` Angus Robinson
2021-07-14 8:32 ` Michael Kjörling
2021-07-14 9:07 ` Lars Brinkhoff
2021-07-14 14:09 ` Larry McVoy
2021-07-14 14:54 ` Warner Losh
2021-07-14 15:06 ` Richard Salz
2021-07-14 15:37 ` Steve Nickolas
2021-07-14 10:09 ` Tom Ivar Helbekkmo via TUHS
2021-07-14 10:39 ` arnold
2021-07-14 17:21 ` Lyndon Nerenberg (VE7TFX/VE6BBM)
2021-07-14 17:32 ` Richard Salz
2021-07-14 15:01 ` Clem Cole
2021-07-14 17:40 ` [TUHS] [COFF] " Theodore Y. Ts'o
2021-07-14 17:50 ` Larry McVoy
2021-07-14 18:28 ` Clem Cole
2021-07-14 11:49 ` [TUHS] " Andy Kosela
2021-07-14 15:48 ` Theodore Y. Ts'o
2021-07-16 1:35 ` Dave Horsfall
2021-07-16 2:33 ` risner
2021-07-16 4:25 ` Theodore Y. Ts'o
2021-07-16 5:51 ` Bakul Shah
2021-07-16 13:00 ` Theodore Y. Ts'o
2021-07-16 13:56 ` Larry McVoy
2021-07-16 14:40 ` Clem Cole
2021-07-16 15:44 ` Theodore Y. Ts'o [this message]
2021-07-16 16:11 ` Bakul Shah
2021-07-16 19:07 ` Kevin Bowling
2021-07-16 20:17 ` Clem Cole
2021-07-16 20:24 ` Richard Salz
2021-07-18 13:13 ` arnold
2021-07-18 13:23 ` Richard Salz
2021-07-18 13:43 ` [TUHS] MtXinu calendar (was Re: 386BSD released) Al Kossow
2021-07-18 13:51 ` Al Kossow
2021-07-18 16:44 ` Al Kossow
2021-07-18 17:38 ` John Cowan
2021-07-18 18:35 ` Bakul Shah
2021-07-19 3:06 ` Dan Stromberg
2021-07-18 19:00 ` arnold
2021-07-18 21:48 ` Deborah Scherrer
2021-07-18 20:06 ` Lyle Bickley
2021-07-14 21:37 ` [TUHS] 386BSD released Bakul Shah
2021-07-16 21:22 ` Dave Horsfall
2021-07-15 2:21 Douglas McIlroy
2021-07-15 2:41 ` Adam Thornton
2021-07-15 15:07 ` Clem Cole
2021-07-16 17:30 Nelson H. F. Beebe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YPGpWh60wFePL1en@mit.edu \
--to=tytso@mit.edu \
--cc=bakul@iitbombay.org \
--cc=clemc@ccc.com \
--cc=tuhs@tuhs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).