Subject: Linux-Development Digest #735
From: Digestifier <Linux-Development-Request@senator-bedfellow.MIT.EDU>
To: Linux-Development@senator-bedfellow.MIT.EDU
Reply-To: Linux-Development@senator-bedfellow.MIT.EDU
Date:     Thu, 19 May 94 20:13:16 EDT

Linux-Development Digest #735, Volume #1         Thu, 19 May 94 20:13:16 EDT

Contents:
  Re: SIGHUP - Deep Kernal Guts question! (Sean Puckett)
  Re: 32-bit Novell desktop OS combines Unix, DOS 7 (David Konerding)
  Re: Anybody working on BSD dump porting? (David E. Fox)
  Re: strftime() bug (Winnie-the-Pooh)
  Yes there is a sequence problem with net 3 (Bradley E. Smith)
  Cache-optimizing page allocation  (Harm Hanemaaijer)

----------------------------------------------------------------------------

From: nate@loreli.ftl.fl.us (Sean Puckett)
Subject: Re: SIGHUP - Deep Kernal Guts question!
Date: Thu, 19 May 1994 01:42:01 GMT

All this stuff about the POSIX standard and manuals is very enlightening,
thank you gentlemen very much.  However, it simply raises more questions.
Now that we know why the kernal doesn't send a HUP to running apps, the
answer being 'because the specs say it', we now must ask why the specs
don't say to send a HUP to running apps.

Why is it in the specs?

Some lame reason like "because it has always been done that way" is not
acceptable.

We are programmers here, and we have a problem that needs
to be resolved.  If there are no hard, fast reasons why it should NOT
send a HUP signal (which is easily ignored, mind you) to all processes
owned by a terminal on loss of carrier, then I will simply change the kernal
to do so.

So again, I ask, why?

What will break if I change the kernel?


--
                       ..:: nate@loreli.ftl.fl.us ::..
             ..:: Sean Puckett - Albino Frog Software, Inc. ::..
          ..:: The Right Reverend Aural Hardly, MSK, BoC, FCoC ::..

------------------------------

From: rafael@cse.ucsc.edu (David Konerding)
Subject: Re: 32-bit Novell desktop OS combines Unix, DOS 7
Date: 19 May 1994 16:00:21 GMT

Eric Kimminau (ekimmina@pms709.pms.ford.com) wrote:
: Reprinted verbatim from an article by Anne Knowles in the May 16th,
: 1994 issue of PCWeek Magazine without permission of Ziff publishing
: Corp. The article appeared on page 1, with a continuation on page 14.

: 32-bit Novell desktop OS combines Unix, DOS 7
: by Anne Knowles

: Novell Inc. is targeting July for the launch of its 32bit multitasking
: desktop operating system called Corsair. Tentatively priced at $99.00,
: the operating system combines Novell DOS7 with a kerrnel based on
: Linux 1.0, a Unix clone for Intel Corp. PCs that is freely available
: over the Internet, according to a Novell internal document examined by
: PCWeek.

: A Novell spokeswoman cautioned that there is no set release date for
: Corsair. And since corporate buyers must already sift through several
: commercial operating systems, some users questioned where the new
: software would find a home.


        I'll tell you this: all of this about "Expose" that we're reading is
market hype, *not* reality.  This operating system is bogus- and nothing
approaching the advertised power it claims to have will be released from Novell
for at least a couple years, if ever.

        Why is PCWeek printing this?  You'd think that computer magazines wouldn't
fall for this.  Wait- what am I saying?  When has a computer magazine *ever* failed
to talk about vaporware as if it's right around the corner?

        
--
   O~_   ______ David Konerding (University of California, Santa Cruz)
  c/ /'    ____ rafael@cse.ucsc.edu
 ( ) \( )    __ rafael@cats.ucsc.edu


------------------------------

Crossposted-To: comp.os.linux.help
From: root@belvedere.sbay.org (David E. Fox)
Subject: Re: Anybody working on BSD dump porting?
Date: Wed, 18 May 1994 21:00:39 GMT

Rob Janssen (rob@pe1chl.ampr.org) wrote:

: Because of the many filesystems that Linux supports, trying to port
: something like "dump" is not simple, and IMHO not a good idea.

Well, ext2fs is fairly common these days.

But what advantages does 'dump' have that 'tar' or 'cpio' do not?  It does
seem to offer extra flexibility because you can have dump levels.  I've
only used 'dump' on 386BSD once.  Lacking a tape drive that would work
under 386BSD, I had to use floppies.  Getting the syntax correct for
floppy volumes was a bit difficult at first, and dump spawned a new
process for each floppy.  At or near volume 20, my load average
skyrocketed (I had 4 megs ram + 5 megs swap at the time) and the system
crashed -- a clear sign that the system had run out of VM. :(

: -------------------------------------------------------------------------
: | Rob Janssen                | AMPRnet:   rob@pe1chl.ampr.org           |
: | e-mail: pe1chl@rabo.nl     | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU     |
: -------------------------------------------------------------------------
-- 
David Fox                       root@belvedere.sbay.org
5479 Castle Manor Drive
San Jose, CA 95129              Thanks for letting me change
408/253-7992                    magnetic patterns on your hard disk.

------------------------------

From: haijo@cs.rug.nl (Winnie-the-Pooh)
Subject: Re: strftime() bug
Date: 19 May 1994 16:43:36 +0200

brandes@cs.tu-berlin.de (Olaf Brandes) writes:

&Hi folks,

&I've come across the phenomen that strftime()'s %j outputs the same value
&as the struct tm member tm_yday.
&According to POSIX, the %j output value should be in the range from 001
&to 366. Actually, values 000 to 365 are returned.

&I guess there aren't too many applications using %j. But if there are
&some, I donn't know what's going to happen.

Everything will happen a day to late. ;)


Winnie-the-Pooh
-- 
finger haijo@midas.cs.rug.nl for pgp public key.

------------------------------

From: brad@bradley.bradley.edu (Bradley E. Smith)
Subject: Yes there is a sequence problem with net 3
Date: 19 May 1994 11:05:16 -0500

included is my sniffer output of me trying to run 'pine' from my
cisco terminal to my linux box....I am looking at the tcp.c code
but if someone with more knowledge can give me some insights, I'd
be grateful.

--
brad smith, brad@bradley.edu
--

Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 1




- - - - - - - - - - - - - - - - Frame 1 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<0D0A>
Telnet:


- - - - - - - - - - - - - - - - Frame 2 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938622024
TCP:  Acknowledgment number = 3659176514
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = E47C (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 3 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<0D0A>
Telnet:


- - - - - - - - - - - - - - - - Frame 4 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176514
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 2


TCP:  Acknowledgment number = 938622027
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1169
TCP:  Checksum = FFE8 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 5 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<1B>[;H<1B>[2J<1B>[7m  PINE 3.89   MAIN MENU                          Folder...
Telnet:


- - - - - - - - - - - - - - - - Frame 6 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176514
TCP:  Acknowledgment number = 938622128
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1068
TCP:  Checksum = FFE8 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 7 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 3


Telnet:
Telnet:<1B>[4;5H   ?     HELP               -  Get help using Pine<1B>[5;5H<1B>[6;5...
Telnet:


- - - - - - - - - - - - - - - - Frame 8 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176514
TCP:  Acknowledgment number = 938622917
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1460
TCP:  Checksum = FB4B (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 9 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<1B>[7m<1B>[1;2H*<1B>[22;1H
Telnet:


- - - - - - - - - - - - - - - - Frame 10 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176514
TCP:  Acknowledgment number = 938622935
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 4


TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1442
TCP:  Checksum = FB4B (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 11 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<1B>[1;2H <1B>[m<1B>[7m<0808>  PINE 3.89   MAIN MENU...
Telnet:


- - - - - - - - - - - - - - - - Frame 12 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176514
TCP:  Acknowledgment number = 938623110
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1267
TCP:  Checksum = FB4B (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 13 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:l
Telnet:


- - - - - - - - - - - - - - - - Frame 14 - - - - - - - - - - - - - - - - -

Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 5



TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623110
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = E03D (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 15 - - - - - - - - - - - - - - - - -


Telnet:----- Telnet data -----
Telnet:
Telnet:<1B>[;H<1B>[2J<1B>[7m  PINE 3.89   FOLDER LIST                           Fol...
Telnet:


- - - - - - - - - - - - - - - - Frame 16 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 6




- - - - - - - - - - - - - - - - Frame 17 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 18 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 19 - - - - - - - - - - - - - - - - -


Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 7


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 20 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 21 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 8


TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 22 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 23 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 9


TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 24 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 25 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 10


TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 26 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 27 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 28 - - - - - - - - - - - - - - - - -

Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 11



TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 29 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 23 (Telnet)
TCP:  Destination port = 42513
TCP:  Sequence number = 938623210
TCP:  Acknowledgment number = 3659176515
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 8192
TCP:  Checksum = DFD9 (correct)
TCP:  No TCP options
TCP:  


- - - - - - - - - - - - - - - - Frame 30 - - - - - - - - - - - - - - - - -


TCP:  ----- TCP header -----
TCP:  
TCP:  Source port = 42513
TCP:  Destination port = 23 (Telnet)
Sniffer Network Analyzer data from 19-May-94 at 10:24:22, unsaved capture data, Page 12


TCP:  Sequence number = 3659176515
TCP:  Acknowledgment number = 938623211
TCP:  Data offset = 20 bytes
TCP:  Flags = 10
TCP:  ..0. .... = (No urgent pointer)
TCP:  ...1 .... = Acknowledgment
TCP:  .... 0... = (No push)
TCP:  .... .0.. = (No reset)
TCP:  .... ..0. = (No SYN)
TCP:  .... ...0 = (No FIN)
TCP:  Window = 1166
TCP:  Checksum = FB4A (correct)
TCP:  No TCP options
TCP:  
-- 
Bradley Smith                    brad@bradley.edu ---  309-677-2337
Network & Technical Services @ Bradley University, Peoria, IL

"It's amazing how much scrap metal you get from 4 cans of beer"

------------------------------

Crossposted-To: comp.sys.ibm.pc.misc
From: hhanemaa@cs.ruu.nl (Harm Hanemaaijer)
Subject: Cache-optimizing page allocation 
Date: Thu, 19 May 1994 14:09:25 GMT

>In article <WAYNE.94May15141143@backbone.uucp>,
>Wayne Schlitt <wayne@cse.unl.edu> wrote:
>>Direct mapped caches tend to have the same performance as 2-way
>>set-associative caches of half the size.  So, direct map caches infate
>>the memory footprint by a factor of 2 or so.
>
>Some studies have shown that it's possibel to greatly reduce this
>factor (on a physically mapped cache) by moving the pages that
>trashes.
>
>It does require some bookkeeping thought, and I don't think any OS
>have implemented this yet.

I have been thinking how nice it would be to do this in Linux...
It would be very beneficial (even with proper write-back caches).

One thing I did was write an optimizing malloc that allocates are large
pool of pages and then mmap's pages together for allocation requests,
keeping track of how many pages have been used for every cache page position
(there are 64 with a 256K direct-mapped cache and 4K pagesize).

This had the effect of improving the performance of one animation type
program, copying two 64K virtual framebuffer each frame, from an average
of 51 frames/s (47-53 roughly) to about 62 frames/s (57-70 roughly). The
program did do some other memory trashing each frame also.

Only the two 64K buffers where allocated in this way, all the other
code/data/stack pages were subject to conflicts (which explains the
significant variation, with a high of about 70).

I think should not be too difficult to implement something like this at
the kernel level with a simple per-process hit count array. Dynamically
detecting trashing pages sounds rather more difficult though. 

hhanemaa@cs.ruu.nl


------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.development) via:

    Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Development Digest
******************************
